WEBVTT 1 00:00:00.000 --> 00:00:03.879 All right, so we're diving into some excerpts from mining 2 00:00:03.960 --> 00:00:07.160 social media finding stories in Internet data today. 3 00:00:07.240 --> 00:00:08.160 Yeah, this should be fun. 4 00:00:08.359 --> 00:00:11.240 It looks like we've got a lot to unpack, from 5 00:00:11.919 --> 00:00:16.719 bot detection to ethical data scraping to even like real 6 00:00:16.800 --> 00:00:19.480 reddit data on vaccinations. 7 00:00:18.719 --> 00:00:19.320 That's to cover. 8 00:00:19.559 --> 00:00:23.679 Yeah, so where do we want to begin? I guess 9 00:00:23.719 --> 00:00:26.199 like bots on Twitter. Sure, they kind of give me 10 00:00:26.239 --> 00:00:27.160 the creeps honestly. 11 00:00:27.280 --> 00:00:28.640 Yeah, they're definitely concerned. 12 00:00:28.719 --> 00:00:30.600 How big of a problem are they really? Like, are 13 00:00:30.679 --> 00:00:32.719 they actually swaying public opinion? 14 00:00:32.960 --> 00:00:37.000 Well, they can definitely amplify certain narratives and manipulate trends, 15 00:00:37.240 --> 00:00:40.759 you know, and even like so discord among real users. 16 00:00:40.479 --> 00:00:41.920 Like a digital echo chamber you. 17 00:00:42.159 --> 00:00:45.840 Yeah, exactly. It creates that illusion of like widespread support 18 00:00:45.920 --> 00:00:48.280 or opposition to an idea when it might not really 19 00:00:48.320 --> 00:00:48.719 be there. 20 00:00:49.000 --> 00:00:51.000 So how can we spot these bots? 21 00:00:51.079 --> 00:00:52.799 Well, one of the easiest ways is to look at 22 00:00:52.799 --> 00:00:54.520 their tweeting frequency. 23 00:00:54.159 --> 00:00:55.799 Like how often they tweet exactly. 24 00:00:56.079 --> 00:00:59.039 Bots can tweet way more than any human possibly could. 25 00:00:59.280 --> 00:01:01.880 Ah, they're like tweeting machines pretty much. 26 00:01:02.399 --> 00:01:04.760 There's an example in the book about this account that's 27 00:01:04.760 --> 00:01:05.920 cinever sets one hundred. 28 00:01:06.040 --> 00:01:07.079 Okay, what about it. 29 00:01:07.079 --> 00:01:10.200 It was flagged for tweeting over seventy times in a 30 00:01:10.239 --> 00:01:10.719 single day. 31 00:01:10.879 --> 00:01:12.439 Wow, that's a lot. 32 00:01:12.680 --> 00:01:14.239 Yeah, no human could keep up with that. 33 00:01:14.400 --> 00:01:16.439 So once you've spotted a potential bought. 34 00:01:16.439 --> 00:01:19.359 What then, well, believe it or not, you can actually 35 00:01:19.359 --> 00:01:20.920 do a lot with Google Sheets. 36 00:01:21.120 --> 00:01:22.519 Really Google Sheet. 37 00:01:22.640 --> 00:01:24.959 Yeah, I know it sounds basic, but you can get 38 00:01:24.959 --> 00:01:27.239 some pretty interesting insights if you know how to use it. 39 00:01:27.359 --> 00:01:30.239 Humh okay, what's the catch? 40 00:01:30.680 --> 00:01:33.799 Well, you have to make sure your data is formatted correctly, 41 00:01:34.599 --> 00:01:37.159 you know, Like, so Google Sheets can tell the difference 42 00:01:37.200 --> 00:01:40.599 between like text numbers and dates, right right. Then you 43 00:01:40.599 --> 00:01:43.560 can use pivot tables to summarize your data, Like, so 44 00:01:43.640 --> 00:01:45.400 you want to see the daily tweet counts for a 45 00:01:45.439 --> 00:01:48.159 suspected bot account, A pivot table can do that. 46 00:01:48.719 --> 00:01:50.680 Interesting, So you're telling me, I can use the same 47 00:01:50.760 --> 00:01:53.640 tool I use to track my Grotery budget to analyze 48 00:01:53.680 --> 00:01:55.159 bot activity exactly. 49 00:01:55.680 --> 00:01:57.920 And you can go even further with formulas. They're like 50 00:01:57.959 --> 00:01:59.519 Google Sheets version of code. 51 00:01:59.640 --> 00:02:03.400 Okay, so it's like you're giving Google Sheets instructions to 52 00:02:03.439 --> 00:02:05.280 manipulate the data exactly. 53 00:02:06.319 --> 00:02:10.479 There's one formula look up that lets you combine data 54 00:02:10.520 --> 00:02:14.759 from different sets by finding matching values. And then there's 55 00:02:14.840 --> 00:02:17.360 ipher which helps you handle errors. 56 00:02:17.439 --> 00:02:21.800 So it's like basic coding, but in a spreadsheet. I'm intrigued. 57 00:02:21.919 --> 00:02:23.400 Yeah, it's pretty powerful stuff. 58 00:02:23.439 --> 00:02:25.400 But what if you want to get even more advanced. 59 00:02:25.400 --> 00:02:27.360 I've heard a lot about Python being the go to 60 00:02:27.520 --> 00:02:28.919 language for data analysis. 61 00:02:29.080 --> 00:02:32.400 Yeah, Python is definitely the next level, especially for handling 62 00:02:32.479 --> 00:02:33.400 large data sets. 63 00:02:33.599 --> 00:02:37.199 All right, so Python large data sets. It sounds intimidating, It's. 64 00:02:37.000 --> 00:02:39.199 Not as bad as it sounds. One important thing to 65 00:02:39.280 --> 00:02:42.280 understand is the concept of virtual environments. 66 00:02:42.479 --> 00:02:44.199 Virtual environments, like, what are those? 67 00:02:44.319 --> 00:02:48.080 They basically help you manage different libraries without causing conflicts. 68 00:02:48.199 --> 00:02:48.759 Libraries. 69 00:02:48.840 --> 00:02:51.840 Yeah, libraries are basically collections of pre written code for 70 00:02:51.879 --> 00:02:54.639 specific tasks, kind of like specialized toolkits. 71 00:02:54.800 --> 00:02:59.039 Virtual environments are like separate workspaces for your Python projects. 72 00:02:59.080 --> 00:03:00.759 Exactly makes sense. 73 00:03:00.960 --> 00:03:03.199 Then what well, once you've got your environment set up, 74 00:03:03.400 --> 00:03:05.960 you can use Jupiter notebook to write and run your 75 00:03:06.000 --> 00:03:06.759 Python code. 76 00:03:06.879 --> 00:03:10.639 Jupiter Notebook, got it. And what about pandas. I've heard 77 00:03:10.639 --> 00:03:13.159 that name thrown around a lot in data analysis circles. 78 00:03:13.319 --> 00:03:16.360 PANDAS is a game changer, especially for social media data. 79 00:03:16.439 --> 00:03:20.400 It's a library that's specifically designed for handling those massive 80 00:03:20.439 --> 00:03:21.000 data sets. 81 00:03:21.080 --> 00:03:22.759 So it helps you make sense of all that data. 82 00:03:22.919 --> 00:03:26.599 Yeah, you can clean it, manipulate it, analyze it. It's 83 00:03:26.639 --> 00:03:29.159 a must have for any serious data analyst. 84 00:03:29.560 --> 00:03:34.719 Okay, so we've talked bots, Google sheets, Python, pandas. What 85 00:03:34.800 --> 00:03:38.039 about that Reddit data you mentioned, the stuff about vaccinations. 86 00:03:38.199 --> 00:03:39.919 Yeah, we can dive into that next. We'll use that 87 00:03:39.960 --> 00:03:42.439 as a case study. There's tons of data from Reddit 88 00:03:42.520 --> 00:03:46.120 thanks to this guy, Jason Baumgartner, who's like a data archivist. 89 00:03:46.479 --> 00:03:48.639 Cool, So what specifically are we going to look at. 90 00:03:48.680 --> 00:03:50.919 We'll focus on the rask science subreddit. We can see 91 00:03:50.919 --> 00:03:53.000 how people are talking about vaccinations online. 92 00:03:53.120 --> 00:03:56.199 Sounds fascinating. And how do we even begin to analyze 93 00:03:56.199 --> 00:03:57.120 all of that data? 94 00:03:57.159 --> 00:04:01.039 We'll use pandas. In Jupiter Notebook, there were these handy 95 00:04:01.120 --> 00:04:04.639 methods like dot head, dot columns, and dot ilock that 96 00:04:04.719 --> 00:04:06.199 help you get a feel for the data. 97 00:04:06.280 --> 00:04:08.560 So like what's in there, the names of the columns, 98 00:04:08.599 --> 00:04:11.479 how to select specific data points exactly. 99 00:04:12.120 --> 00:04:14.639 But fair warning, we're probably going to run into some 100 00:04:14.719 --> 00:04:16.439 missing values in the data. 101 00:04:16.560 --> 00:04:18.879 Missing values like what Yeah. 102 00:04:18.720 --> 00:04:21.879 Like netive or nan entries. They can miss things up 103 00:04:21.879 --> 00:04:22.720 if you're not careful. 104 00:04:22.839 --> 00:04:23.959 So what do you do about them. 105 00:04:24.199 --> 00:04:27.480 You can either remove those roads entirely with dot dropna 106 00:04:27.959 --> 00:04:30.800 or replace them with something else using dot filma. It 107 00:04:30.839 --> 00:04:32.519 really depends on what you're trying to find out. 108 00:04:33.000 --> 00:04:36.839 So no one size fits all solution, got it. So 109 00:04:37.160 --> 00:04:41.000 how do we actually go about analyzing these Reddit conversations 110 00:04:41.040 --> 00:04:42.199 about vaccinations. 111 00:04:42.439 --> 00:04:44.079 Well, first we need to figure out what we're trying 112 00:04:44.120 --> 00:04:46.519 to understand, you know, like are we trying to gauge 113 00:04:46.560 --> 00:04:50.199 overall sentiment or are we looking for specific themes or patterns. 114 00:04:51.319 --> 00:04:54.759 I'm really interested in how engaged people are in these discussions, 115 00:04:54.839 --> 00:04:58.160 like are they generally supportive or hesitant? Are there any 116 00:04:58.199 --> 00:05:01.839 common arguments or concerns keep pupping up perfect We. 117 00:05:01.800 --> 00:05:03.839 Can definitely look into that. We'll need to look at 118 00:05:03.839 --> 00:05:06.160 both the content of the posts and things like up votes. 119 00:05:05.959 --> 00:05:07.759 And comments, right, so we're not just looking at what 120 00:05:07.800 --> 00:05:11.279 people are saying, but also how others are reacting. 121 00:05:10.839 --> 00:05:13.480 To it exactly. And that's where the idea of central 122 00:05:13.519 --> 00:05:16.680 tendency comes in. We can use statistical measures like the 123 00:05:16.759 --> 00:05:19.319 mean and median to get a sense of the average engagement. 124 00:05:19.639 --> 00:05:22.600 So like if a lot of pro vaccination posts have 125 00:05:22.720 --> 00:05:25.279 a ton of up votes, that might suggest there's a 126 00:05:25.319 --> 00:05:27.000 lot of support for that viewpoint. 127 00:05:27.240 --> 00:05:29.959 Yeah, it could, but we have to be careful about 128 00:05:30.040 --> 00:05:33.439 jumping to conclusions. There might be other things at play. Right. 129 00:05:33.600 --> 00:05:37.279 Correlation doesn't equal causation, so we can't just assume that 130 00:05:37.399 --> 00:05:40.040 upvotes equal agreement, right exactly. 131 00:05:40.240 --> 00:05:42.560 That's why it's so important to look at multiple factors 132 00:05:42.800 --> 00:05:44.160 and to consider the context. 133 00:05:44.399 --> 00:05:47.680 Okay, makes sense. But before we get too deep into analysis, 134 00:05:47.720 --> 00:05:50.759 I'm guessing we need to narrow down this massive Reddit 135 00:05:50.839 --> 00:05:53.639 data set to just the stuff about vaccinations. 136 00:05:53.720 --> 00:05:56.079 Right, Yeah, we don't want to waste time sifting through 137 00:05:56.160 --> 00:05:57.279 irrelevant posts. 138 00:05:57.439 --> 00:05:58.319 So how do we do that? 139 00:05:58.639 --> 00:06:01.720 Well, pandas is great for filtering data. We can create 140 00:06:01.759 --> 00:06:05.040 a new data frame that only includes posts with certain 141 00:06:05.120 --> 00:06:07.600 keywords related to vaccinations. 142 00:06:07.079 --> 00:06:09.399 Like a supercharged search function. 143 00:06:09.240 --> 00:06:11.759 Basically pretty much. And then we can start looking at 144 00:06:11.800 --> 00:06:13.240 those engagement metrics. 145 00:06:13.040 --> 00:06:15.959 Right, all those upvotes and comments. How do we make 146 00:06:16.000 --> 00:06:16.839 sense of all of that? 147 00:06:17.160 --> 00:06:20.519 We can combine those columns into a new metric like 148 00:06:20.600 --> 00:06:25.079 combined engagement, and then calculate the average using dot mean. 149 00:06:25.600 --> 00:06:28.000 We can also use dusk gribe to get a better 150 00:06:28.120 --> 00:06:31.079 understanding of the distribution of that engagement. 151 00:06:30.800 --> 00:06:33.199 So we might see that some posts get way more 152 00:06:33.199 --> 00:06:36.720 engagement than others, even if the average is relatively consistent. 153 00:06:36.959 --> 00:06:40.000 Exactly, and those outliers can tell us a lot about 154 00:06:40.000 --> 00:06:41.480 what's really driving the conversation. 155 00:06:41.600 --> 00:06:43.879 Okay, I'm starting to see how this all comes together. Now, 156 00:06:44.240 --> 00:06:46.480 can we switch gears and talk about Facebook for a second. 157 00:06:46.480 --> 00:06:48.759 I know, a little off topic, but sure I get 158 00:06:48.759 --> 00:06:50.959 a little creaked out by how much data Facebook collects 159 00:06:51.000 --> 00:06:51.279 on us. 160 00:06:51.399 --> 00:06:52.279 Yeah, it's a lot. 161 00:06:52.360 --> 00:06:55.879 But I did hear you can download an archive of 162 00:06:56.199 --> 00:06:57.399 your user data. 163 00:06:57.600 --> 00:06:58.040 That's right. 164 00:06:58.079 --> 00:06:59.279 What kind of stuff is in there? Oh? 165 00:06:59.319 --> 00:07:02.240 Pretty much? Every thing? Your posts, your interactions, the ads 166 00:07:02.240 --> 00:07:02.839 you've clicked on. 167 00:07:02.920 --> 00:07:04.639 Wait, they track what ads I click on? Yep. 168 00:07:04.879 --> 00:07:05.319 Everything. 169 00:07:05.639 --> 00:07:08.959 That's kind of creepy but also kind of fascinating. Yeah, 170 00:07:09.040 --> 00:07:10.759 I'd love to know what kind of insights I could 171 00:07:10.759 --> 00:07:11.680 get from all that data. 172 00:07:11.759 --> 00:07:13.319 Well, that's where webscraping comes in. 173 00:07:13.399 --> 00:07:14.839 Web scraping what is that? 174 00:07:15.240 --> 00:07:19.560 It's basically a way to extract specific information from websites 175 00:07:20.160 --> 00:07:20.879 using code. 176 00:07:21.240 --> 00:07:24.079 So you're telling me I can use code to dig 177 00:07:24.120 --> 00:07:26.560 through my Facebook data and find out what they know 178 00:07:26.639 --> 00:07:27.000 about me. 179 00:07:27.319 --> 00:07:28.319 Yeah, pretty much. 180 00:07:28.399 --> 00:07:30.720 That's both amazing and terrifying. 181 00:07:30.800 --> 00:07:31.120 It is. 182 00:07:31.439 --> 00:07:34.040 But before I go on a Facebook data mining spree. 183 00:07:34.079 --> 00:07:36.959 I imagine there are some ethical considerations here, right, Oh? 184 00:07:36.959 --> 00:07:40.800 Absolutely. One big one is the robot's exclusion protocol. 185 00:07:40.639 --> 00:07:42.680 Robots exclusion protocol. What's that? 186 00:07:42.920 --> 00:07:45.800 It's basically a set of rules that websites use to 187 00:07:45.839 --> 00:07:49.199 tell web robots, which are like automated programs that browse 188 00:07:49.240 --> 00:07:51.360 the web, which parts of the site they can and 189 00:07:51.399 --> 00:07:52.079 can't access. 190 00:07:52.480 --> 00:07:54.920 Okay, so it's like a digital do not enter sign 191 00:07:55.040 --> 00:07:56.600 for bots pretty much. 192 00:07:57.040 --> 00:07:59.800 And these rules are outlined in a file called robots 193 00:07:59.800 --> 00:08:02.319 dot txt. Every website has one. 194 00:08:02.240 --> 00:08:04.120 So if I want to scrape data from a website 195 00:08:04.120 --> 00:08:06.800 I need to check their robots dot txt file first 196 00:08:06.800 --> 00:08:08.839 to make sure I'm not breaking any rules exactly. 197 00:08:09.160 --> 00:08:11.120 It's about being respectful of those boundaries. 198 00:08:11.160 --> 00:08:13.319 Makes sense. So how does this apply to scraping my 199 00:08:13.399 --> 00:08:14.319 Facebook data? 200 00:08:14.439 --> 00:08:19.120 Well, Facebook's robots dot txt file will likely restrict scraping 201 00:08:19.240 --> 00:08:22.759 certain types of data like user profiles or private messages. 202 00:08:22.920 --> 00:08:26.279 So I can't just go willy nilly scraping everything on Facebook. 203 00:08:26.439 --> 00:08:28.199 Nop. You gotta play by the rule. 204 00:08:28.199 --> 00:08:31.639 Okay, I get it FICX first, But assuming I am 205 00:08:31.759 --> 00:08:34.759 following the rules, how do I actually do this web 206 00:08:34.799 --> 00:08:36.720 scraping thing? What kind of tools do I need. 207 00:08:36.799 --> 00:08:39.559 Python is great for webscraping, especially when you combine it 208 00:08:39.600 --> 00:08:41.399 with a library called beautiful Soup. 209 00:08:41.519 --> 00:08:44.240 Beautiful Soup huh, interesting name. What does it do? 210 00:08:44.519 --> 00:08:47.759 Beautiful Soup helps you parse HTML content, which is the 211 00:08:47.799 --> 00:08:51.440 code that structures web pages, and extract the data you need. 212 00:08:51.639 --> 00:08:54.519 So it's like a digital detective, sifting through all that 213 00:08:54.600 --> 00:08:56.240 code and finding the clues I'm looking for. 214 00:08:56.440 --> 00:08:59.279 Exactly. It helps you make sense of the messy world 215 00:08:59.320 --> 00:08:59.919 of web data. 216 00:09:00.480 --> 00:09:03.360 So I could use beautiful Soup to, say, extract all 217 00:09:03.399 --> 00:09:05.519 the ads I've clicked on from my Facebook archive. 218 00:09:05.399 --> 00:09:07.720 Exactly, and then you can analyze that data to see 219 00:09:07.720 --> 00:09:10.919 what kind of patterns emerge. You might be surprised by 220 00:09:11.000 --> 00:09:11.519 what you find. 221 00:09:11.799 --> 00:09:14.840 This is really opening up a whole new world of possibilities. 222 00:09:15.080 --> 00:09:18.639 But I'm realizing the data itself is only half the story, right, 223 00:09:19.159 --> 00:09:21.320 is what we do with it that really matters. 224 00:09:21.440 --> 00:09:24.879 Absolutely, The real magic happens when we start interpreting that data, 225 00:09:25.360 --> 00:09:28.120 drawing conclusions, and telling stories with it. 226 00:09:28.279 --> 00:09:31.080 So data analysis is more than just a technical skill. 227 00:09:31.080 --> 00:09:33.200 It's a form of storytelling. 228 00:09:32.639 --> 00:09:36.440 Exactly, and those stories have the power to inform, inspire, 229 00:09:37.000 --> 00:09:38.320 and even change the world. 230 00:09:38.559 --> 00:09:41.240 I'm sold. This is way more exciting than I ever imagine. 231 00:09:41.320 --> 00:09:44.399 Yeah, it's pretty cool stuff. So we're to next. We've 232 00:09:44.399 --> 00:09:45.799 got a lot of ground to cover still. 233 00:09:45.919 --> 00:09:49.279 Hmm, well before we move on to something completely different, 234 00:09:49.320 --> 00:09:50.919 I'm kind of curious about Wikipedia. 235 00:09:51.200 --> 00:09:52.559 Okay, what about it? 236 00:09:52.559 --> 00:09:55.080 It's such a massive source of information, right, like a 237 00:09:55.240 --> 00:09:58.679 giant online encyclopedia it is. I bet there are some 238 00:09:58.919 --> 00:10:01.840 amazing stories and all that data. You're right there are, 239 00:10:02.000 --> 00:10:04.919 But I imagine it's also quite challenging to scrape data 240 00:10:04.919 --> 00:10:05.879 from Wikipedia. 241 00:10:06.039 --> 00:10:08.759 It can be. It's a very dynamic website, constantly being 242 00:10:08.840 --> 00:10:11.519 updated and edited by volunteers all over the world. 243 00:10:11.759 --> 00:10:15.120 So how do you even approach scraping data from something 244 00:10:15.200 --> 00:10:16.039 like Wikipedia. 245 00:10:16.600 --> 00:10:20.000 Patients and the right tools are key, and a good 246 00:10:20.120 --> 00:10:22.759 understanding of Wikipedia structure and how it works. 247 00:10:23.000 --> 00:10:26.639 Okay, so what if let's say I wanted to compile 248 00:10:26.720 --> 00:10:30.159 a list of all the women computer scientists listed on Wikipedia. 249 00:10:30.480 --> 00:10:34.360 That's a great example. Wikipedia has category pages dedicated to 250 00:10:34.399 --> 00:10:38.519 specific topics. You could start with the women computer scientists 251 00:10:38.559 --> 00:10:42.080 category page and use web scraping to extract the names 252 00:10:42.120 --> 00:10:43.759 of all the individuals listed there. 253 00:10:43.759 --> 00:10:45.799 Cool, and I could even grab the links to their 254 00:10:45.799 --> 00:10:47.600 individual Wikipedia pages. 255 00:10:47.600 --> 00:10:50.279 Exactly, And then you could deal deeper into those pages 256 00:10:50.320 --> 00:10:54.240 and extract even more information like their birth dates, nationalities, 257 00:10:54.519 --> 00:10:57.759 areas of expertise. The possibilities are endless. 258 00:10:58.000 --> 00:11:00.200 This is blowing my mind, but I imagine. And there 259 00:11:00.200 --> 00:11:02.480 are some specific considerations we need to keep in mind 260 00:11:02.519 --> 00:11:03.720 when scraping Wikipedia. 261 00:11:03.799 --> 00:11:06.080 Oh absolutely. One of the most important is to be 262 00:11:06.120 --> 00:11:07.919 respectful of their terms of service. 263 00:11:07.679 --> 00:11:10.360 Right, we don't want to crash Wikipedia or anything exactly. 264 00:11:10.919 --> 00:11:14.120 They have guidelines in place to prevent abuse and ensure 265 00:11:14.120 --> 00:11:16.600 that scraping activities don't overload their servers. 266 00:11:16.919 --> 00:11:18.440 So how can we be mindful of that? 267 00:11:18.960 --> 00:11:22.080 Well, one simple technique is to incorporate pauses into your 268 00:11:22.120 --> 00:11:23.039 scraping code. 269 00:11:23.240 --> 00:11:23.720 Pauses. 270 00:11:23.960 --> 00:11:27.679 Yeah, Python has a function called sleep that allows you 271 00:11:27.759 --> 00:11:30.679 to pause the execution of your code for a specified 272 00:11:30.679 --> 00:11:31.320 amount of time. 273 00:11:31.559 --> 00:11:35.039 Ah, so we're giving Wikipedia servers a little breather between 274 00:11:35.080 --> 00:11:36.519 requests exactly. 275 00:11:37.080 --> 00:11:39.639 This can help prevent you from sending too many requests 276 00:11:39.639 --> 00:11:42.600 in quick succession, which could trigger their defenses and get 277 00:11:42.639 --> 00:11:43.840 your IP address blocked. 278 00:11:44.080 --> 00:11:47.440 Okay, so be polite, be patient, and don't overload the system. 279 00:11:47.799 --> 00:11:51.320 Got it. But how do we actually translate our scraping 280 00:11:51.360 --> 00:11:54.759 intentions into Python code that beautiful Soup can understand. 281 00:11:55.399 --> 00:11:58.639 Beautiful Soup makes it pretty intuitive to target specific htmail 282 00:11:58.679 --> 00:12:01.399 elements on a web page. We can use those classes 283 00:12:01.399 --> 00:12:04.320 and ideas we talked about earlier to pinpoint the exact 284 00:12:04.320 --> 00:12:07.480 parts of a Wikipedia page that contain the information we need. 285 00:12:07.600 --> 00:12:10.799 So it's like giving beautiful Soup a treasure map, guiding 286 00:12:10.879 --> 00:12:13.799 it to the exact spot where the digital gold is buried. 287 00:12:14.039 --> 00:12:16.960 I like that analogy. And once you've identified those key 288 00:12:17.120 --> 00:12:21.279 HTML elements, beautiful Soup makes it very easy to extract 289 00:12:21.279 --> 00:12:24.600 the content within them. You can then organize that data 290 00:12:25.159 --> 00:12:27.559 meatly into a spreadsheet for further analysis. 291 00:12:27.639 --> 00:12:30.919 This is amazing. It's like having a superpower that allows 292 00:12:30.960 --> 00:12:33.519 you to unlock the hidden knowledge of the Internet. 293 00:12:33.600 --> 00:12:36.799 It's pretty powerful stuff. But remember the data itself is 294 00:12:36.799 --> 00:12:38.240 only part of the story, right. 295 00:12:38.279 --> 00:12:41.759 We need to analyze it, interpret it, and ultimately tell 296 00:12:41.759 --> 00:12:42.519 a story. 297 00:12:42.240 --> 00:12:46.080 With it exactly. Yeah, and those stories can be incredibly impactful. 298 00:12:46.200 --> 00:12:48.559 This whole deep dive has been eye opening. It's like 299 00:12:48.559 --> 00:12:50.360 I'm seeing the Internet in a whole new light. 300 00:12:50.519 --> 00:12:52.799 I know what you mean. There's so much more to 301 00:12:52.840 --> 00:12:55.120 it than meets the eye. But before we get too 302 00:12:55.159 --> 00:12:58.240 carried away, with webscraping. Let's circle back to Google Sheets 303 00:12:58.279 --> 00:12:58.720 for a moment. 304 00:12:58.840 --> 00:13:01.399 Google Sheets. I thought we were moving on to more 305 00:13:01.440 --> 00:13:02.320 advanced tools. 306 00:13:02.840 --> 00:13:06.679 Google sheets might seem basic, but it's actually quite powerful 307 00:13:06.759 --> 00:13:09.879 for data analysis, especially if you're just starting out. 308 00:13:10.120 --> 00:13:13.320 Hmm, okay, I'm intrigued. What can you do with it? 309 00:13:13.759 --> 00:13:16.320 Well, remember those Twitter bots we talked about earlier. Google 310 00:13:16.360 --> 00:13:18.480 sheets is great for analyzing their activity. 311 00:13:18.600 --> 00:13:21.879 Really, you can track bot behavior in a spreadsheet. 312 00:13:22.000 --> 00:13:25.159 You can. You can even use it to visualize their 313 00:13:25.200 --> 00:13:28.120 tweeting patterns and see if there are any suspicious spikes 314 00:13:28.159 --> 00:13:28.679 or trends. 315 00:13:29.039 --> 00:13:32.000 Wow. I never would have thought of that. So Google 316 00:13:32.000 --> 00:13:35.679 Sheets is like a gateway drug to data analysis. It 317 00:13:35.720 --> 00:13:38.320 helps you get a taste of what's possible before diving 318 00:13:38.360 --> 00:13:40.159 into the more advanced tools. 319 00:13:40.320 --> 00:13:43.799 You could say that, but even experienced data analysts often 320 00:13:43.919 --> 00:13:48.000 use Google Sheets for quick explorations or for creating simple visualization. 321 00:13:48.200 --> 00:13:50.279 Right, So it's not just for beginners exactly. 322 00:13:50.320 --> 00:13:52.399 It's all about choosing the right tool for the job. 323 00:13:53.039 --> 00:13:54.960 Sometimes a simple spreadsheet is all you need. 324 00:13:55.120 --> 00:13:57.600 Okay, I'm starting to see the appeal. What else can 325 00:13:57.639 --> 00:14:00.320 we do with Google Sheets for data analysis? 326 00:14:00.840 --> 00:14:04.000 Well, we could, for instance, analyze the sentiment of those 327 00:14:04.320 --> 00:14:06.480 Reddit posts about vaccinations. 328 00:14:06.759 --> 00:14:09.840 Sentiment you mean, like whether people are generally positive or 329 00:14:09.840 --> 00:14:11.879 negative about vaccinations exactly? 330 00:14:12.120 --> 00:14:14.279 We can look at the words and phrases people using 331 00:14:14.320 --> 00:14:18.440 their posts and use Google Sheets to categorize them as positive, negative, 332 00:14:18.559 --> 00:14:19.120 or neutral. 333 00:14:19.360 --> 00:14:22.840 That sounds incredibly useful. So Google Sheets can help us 334 00:14:22.840 --> 00:14:24.799 go beyond just the numbers and get a sense of 335 00:14:24.799 --> 00:14:27.399 the emotional tone of the conversation precisely. 336 00:14:27.840 --> 00:14:30.519 And we can even use Google Sheets to visualize those 337 00:14:30.559 --> 00:14:33.879 sentiment trends over time. Does either there any shifts or patterns? 338 00:14:34.159 --> 00:14:37.279 I'm starting to see the potential here. Google sheets might 339 00:14:37.320 --> 00:14:39.799 not be as flashy as Python, but it's definitely a 340 00:14:39.879 --> 00:14:41.879 versatile tool for data analysis. 341 00:14:41.960 --> 00:14:43.960 I agree. It's a great place to start for anyone 342 00:14:44.000 --> 00:14:46.240 who's new to data analysis, and it can be a 343 00:14:46.360 --> 00:14:49.159 powerful tool even for experienced analysts. 344 00:14:49.200 --> 00:14:51.679 Okay, I'm officially a Google Sheets convert. 345 00:14:51.759 --> 00:14:54.360 Great. Now, are you ready for something a little more advanced? 346 00:14:54.519 --> 00:14:55.120 Hit me with it. 347 00:14:55.279 --> 00:14:57.559 Let's talk about data analysis for journalists. 348 00:14:57.919 --> 00:15:01.399 Oooh, now this is getting interesting. I've always been fascinated 349 00:15:01.440 --> 00:15:03.799 by the intersection of data and storytelling. 350 00:15:03.879 --> 00:15:07.919 It's a powerful combination. Data journalism is all about using 351 00:15:08.000 --> 00:15:12.679 data to uncover hidden truths, hold the powerful accountable, and 352 00:15:12.720 --> 00:15:13.919 tell stories that matter. 353 00:15:14.320 --> 00:15:18.000 So data journalists are like digital detectives, using data as 354 00:15:18.080 --> 00:15:20.639 clues to solve mysteries and expose wrongdoing. 355 00:15:20.919 --> 00:15:24.120 Exactly, they're using data to dig deeper, to go beyond 356 00:15:24.159 --> 00:15:25.919 the surface, and to find the real story. 357 00:15:26.120 --> 00:15:29.799 This is incredible. Are there any specific examples of how 358 00:15:29.960 --> 00:15:32.639 data journalism is being used to make a difference in 359 00:15:32.679 --> 00:15:33.080 the world. 360 00:15:33.200 --> 00:15:36.679 Oh, there are tons. Data journalists have exposed everything from 361 00:15:36.720 --> 00:15:41.279 corruption and fraud to environmental abuses in human rights violations. 362 00:15:41.320 --> 00:15:44.919 Wow. So data journalism is like a superpower for journalists, 363 00:15:45.039 --> 00:15:48.480 giving them the ability to see things that others can't. 364 00:15:48.600 --> 00:15:51.120 You could say that. And the best part is data 365 00:15:51.159 --> 00:15:55.000 journalism is not limited to large news organizations. Anyone with 366 00:15:55.120 --> 00:15:58.320 access to data and the willingness to learn can use 367 00:15:58.360 --> 00:15:59.000 these techniques. 368 00:15:59.080 --> 00:16:02.399 So it's like a democratizing force, empowering anyone to become 369 00:16:02.399 --> 00:16:04.519 a watchdog and hold the powerful accountable. 370 00:16:04.639 --> 00:16:07.960 Precisely, data journalism is giving a voice to the voiceless 371 00:16:08.279 --> 00:16:10.960 and helping to create a more informed and just society. 372 00:16:11.320 --> 00:16:14.279 This is so inspiring. I'm starting to see the incredible 373 00:16:14.320 --> 00:16:17.120 potential of data analysis to make a real impact in 374 00:16:17.120 --> 00:16:17.519 the world. 375 00:16:17.639 --> 00:16:20.039 I agree. It's a powerful tool for change, and I 376 00:16:20.080 --> 00:16:22.240 think we're only just beginning to scratch the surface of 377 00:16:22.279 --> 00:16:22.960 what's possible. 378 00:16:23.240 --> 00:16:25.240 I can't wait to see what the future holds for 379 00:16:25.320 --> 00:16:28.320 data analysis. It feels like we're on the cusp of 380 00:16:28.320 --> 00:16:30.399 something truly transformative. 381 00:16:30.679 --> 00:16:33.159 I think you're right. The world of data is vast 382 00:16:33.240 --> 00:16:38.559 and ever evolving, and there are endless possibilities for exploration, discovery, 383 00:16:38.639 --> 00:16:39.360 and impact. 384 00:16:39.519 --> 00:16:41.799 I'm ready to dive in. This whole deep dive has 385 00:16:41.799 --> 00:16:45.000 been a revelation. I'm feeling energized and inspired to learn 386 00:16:45.039 --> 00:16:48.039 more and to see what stories I can uncover with data. 387 00:16:48.120 --> 00:16:51.960 That's the spirit, and remember the journey of data analysis 388 00:16:52.080 --> 00:16:56.200 is just as important as the destination. Embrace the challenges, 389 00:16:56.480 --> 00:16:59.919 celebrate the victories, and never stop asking questions. 390 00:17:00.159 --> 00:17:02.159 Wise words. Okay, I think we've covered a lot of 391 00:17:02.200 --> 00:17:04.279 ground for this first part of our deep dive. I'm 392 00:17:04.279 --> 00:17:05.720 excited to see where we go next. 393 00:17:05.799 --> 00:17:08.640 Me too. Let's take a break and come back refresh 394 00:17:08.720 --> 00:17:10.039 for the next part of our exploration. 395 00:17:10.319 --> 00:17:12.279 All right, so before we jump to this part, we 396 00:17:12.279 --> 00:17:14.880 were getting into how AI is becoming a bigger and 397 00:17:14.880 --> 00:17:18.359 bigger part of data analysis. That's super interesting and all, 398 00:17:18.400 --> 00:17:20.200 but I also want to know more about the human 399 00:17:20.319 --> 00:17:23.319 side of this field, Like what kind of person thrives 400 00:17:23.359 --> 00:17:25.920 in a data analysis rule, that's a great question. 401 00:17:26.599 --> 00:17:29.599 You know, the really successful data analysts tend to have 402 00:17:29.640 --> 00:17:33.119 a few things in common, okay, like what, Well, first 403 00:17:33.119 --> 00:17:35.440 of all, they're super curious, makes sense. 404 00:17:35.680 --> 00:17:37.920 Gotta love digging into data. 405 00:17:37.599 --> 00:17:39.880 Right, They're always trying to figure out how things work, 406 00:17:40.200 --> 00:17:43.839 uncover hidden patterns, find those answers to those tough questions. 407 00:17:43.960 --> 00:17:46.359 So it's not just about the numbers. It's about asking 408 00:17:46.400 --> 00:17:48.200 the right question exactly. 409 00:17:47.920 --> 00:17:50.359 Yeah, and figuring out how to use the data to 410 00:17:50.519 --> 00:17:53.799 get those answers. It's a whole process of discovery. 411 00:17:53.319 --> 00:17:55.720 That sounds way more exciting than just staring at a 412 00:17:55.720 --> 00:17:56.640 spreadsheet all day. 413 00:17:56.759 --> 00:17:58.839 Oh, it's definitely not just spreadsheets. You got to be 414 00:17:58.880 --> 00:18:00.039 a creative thinker too. 415 00:18:00.160 --> 00:18:03.680 Really, So there's an artistic side to data analysis. 416 00:18:03.759 --> 00:18:06.680 You could say that you need to be able to 417 00:18:06.759 --> 00:18:10.559 see those connections that others might miss, come up with 418 00:18:10.640 --> 00:18:13.920 new ways to solve problems, and then of course present 419 00:18:13.960 --> 00:18:15.640 your findings in a way that makes sense. 420 00:18:15.920 --> 00:18:18.279 Right. You can't just drown people in data. You got 421 00:18:18.279 --> 00:18:19.160 to tell a story with it. 422 00:18:19.519 --> 00:18:23.400 You got it. Data can be powerful, but it only 423 00:18:23.440 --> 00:18:26.759 really matters if people understand it, and that means turning 424 00:18:26.839 --> 00:18:30.480 it into a story that resonates with them. The best 425 00:18:30.559 --> 00:18:32.880 data analysts they're great storytellers too. 426 00:18:33.039 --> 00:18:35.880 Okay, that makes total sense. Take those raw numbers and 427 00:18:35.920 --> 00:18:38.920 weave them into something that captures people's attention, something that 428 00:18:38.960 --> 00:18:39.599 makes them care. 429 00:18:40.119 --> 00:18:43.240 Right, It's all about finding the human connection within the data. 430 00:18:44.920 --> 00:18:47.599 Oh and that brings me to another important trait, empathy. 431 00:18:48.000 --> 00:18:50.839 Empathy that seems a little unexpected for a field that's 432 00:18:50.839 --> 00:18:51.640 so data driven. 433 00:18:51.920 --> 00:18:55.480 Yeah, it might seem surprising, but it's super important. Remember, 434 00:18:55.599 --> 00:18:58.759 data analysis isn't just about the numbers themselves. It's about 435 00:18:58.839 --> 00:19:03.519 understanding people. Whether you're looking at customer behavior, social media trends, 436 00:19:03.640 --> 00:19:07.680 or even healthcare data, you're ultimately dealing with human experiences. 437 00:19:08.240 --> 00:19:10.240 And if you can put yourself in other people's shoes 438 00:19:10.359 --> 00:19:14.440 understand their perspectives, then you can ask better questions, you know, 439 00:19:14.759 --> 00:19:17.079 and draw more meaningful conclusions from that data. 440 00:19:17.160 --> 00:19:18.960 Okay, that's a really good point. So it's not just 441 00:19:19.000 --> 00:19:22.039 a technical field. It's one where you need to understand people, 442 00:19:22.319 --> 00:19:25.599 connect with them on a human level exactly. It's all 443 00:19:25.640 --> 00:19:29.880 about blending that analytical rigor with empathy, with that human 444 00:19:29.920 --> 00:19:33.440 capacity for understanding, and when you combine those elements you 445 00:19:33.480 --> 00:19:35.200 can achieve some truly amazing things. 446 00:19:35.279 --> 00:19:37.599 Wow. Okay, so we talked about how Python is a 447 00:19:37.640 --> 00:19:42.079 super powerful tool for data analysis, especially when you're dealing 448 00:19:42.079 --> 00:19:44.599 with these massive data sets. Why don't we talk a 449 00:19:44.640 --> 00:19:47.799 little bit more about what makes Python so great for 450 00:19:47.839 --> 00:19:48.640 this kind of work. 451 00:19:48.920 --> 00:19:50.839 Yeah? Sure, Python is super versatile. 452 00:19:51.119 --> 00:19:54.119 So what makes it so popular for data analysis? 453 00:19:54.240 --> 00:19:57.359 Well, first of all, it's got this really clear, readable syntax, 454 00:19:57.880 --> 00:20:01.440 which basically means it's relatively easy to learn and use, 455 00:20:01.839 --> 00:20:02.799 even if you're a beginner. 456 00:20:03.000 --> 00:20:05.119 Okay, so it's not like some super secret code that 457 00:20:05.160 --> 00:20:07.039 only experts can understand. 458 00:20:06.640 --> 00:20:08.640 Not at all. It almost reads like plain English, so 459 00:20:08.640 --> 00:20:10.319 you don't spend all your time trying to figure out 460 00:20:10.319 --> 00:20:10.759 what the code is. 461 00:20:10.799 --> 00:20:12.880 Even saying that's definitely a plus. What else? 462 00:20:13.119 --> 00:20:16.599 Another reason Python is so great for data analysis is 463 00:20:16.640 --> 00:20:20.279 that it's got this huge, active community of developers who 464 00:20:20.359 --> 00:20:24.960 are always creating new libraries and tools specifically for that purpose. 465 00:20:25.240 --> 00:20:28.119 So it's like having a global support system, a whole 466 00:20:28.160 --> 00:20:31.319 team of data enthusiasts ready to help you out. 467 00:20:31.279 --> 00:20:34.440 Exactly you're not alone in this. There are tons of 468 00:20:34.480 --> 00:20:37.039 resources out there to get you started, to help you 469 00:20:37.119 --> 00:20:39.119 tackle any challenge you come across. 470 00:20:39.359 --> 00:20:42.599 That's pretty awesome. And you mentioned libraries earlier. 471 00:20:42.359 --> 00:20:46.480 Right, So libraries they're basically like specialized toolkits for different 472 00:20:46.559 --> 00:20:47.920 data analysis tasks. 473 00:20:48.079 --> 00:20:49.279 So what are some examples. 474 00:20:49.359 --> 00:20:51.519 Okay, So you've got pandas, which we've already talked about 475 00:20:51.519 --> 00:20:56.240 a bit for manipulating and analyzing data. Numb Pi is 476 00:20:56.279 --> 00:20:59.200 great for working with numerical data, and then there are 477 00:20:59.440 --> 00:21:03.039 libraries like map plotlib and seaborne which are all about 478 00:21:03.079 --> 00:21:06.599 creating visualizations. And of course, if you're getting into machine learning, 479 00:21:06.680 --> 00:21:07.680 you've got side kit learn. 480 00:21:07.920 --> 00:21:10.079 Okay, So it's like having a whole arsenal of tools 481 00:21:10.119 --> 00:21:13.440 at your disposal, each one designed for a specific purpose exactly. 482 00:21:13.480 --> 00:21:15.920 And Python makes it super easy to use these libraries, 483 00:21:16.160 --> 00:21:18.000 so you don't have to start from scratch every time. 484 00:21:18.200 --> 00:21:20.759 That sounds incredibly efficient. But going back to something we 485 00:21:20.799 --> 00:21:24.599 touched on earlier, the whole data cleaning and preparation part 486 00:21:24.720 --> 00:21:28.240 of webscriping, Why is that so important? I mean, why 487 00:21:28.279 --> 00:21:30.720 not just dive right into the analysis. 488 00:21:30.079 --> 00:21:34.880 Because real world data it's messy, it's inconsistent, it's not 489 00:21:34.920 --> 00:21:37.880 always perfect. Think of it like cooking. Before you can 490 00:21:37.920 --> 00:21:41.119 make a delicious meal, you got to wash, chop, and 491 00:21:41.240 --> 00:21:44.240 prep those ingredients. It's the same with data. Before you 492 00:21:44.240 --> 00:21:47.119 can draw meaningful insights from it, you got to clean 493 00:21:47.160 --> 00:21:49.279 it up, make sure it's accurate and consistent. 494 00:21:49.519 --> 00:21:52.200 So data cleaning is like laying the groundwork for a 495 00:21:52.359 --> 00:21:53.559 solid analysis. 496 00:21:53.759 --> 00:21:56.400 Exactly. If you start with messy data, you're going to 497 00:21:56.480 --> 00:21:59.759 get messy results, garbage in and garbage out, as they say. 498 00:21:59.599 --> 00:22:02.799 So, what are some common data cleaning tasks? 499 00:22:03.400 --> 00:22:07.319 Well, one common task is handling missing values. Like let's 500 00:22:07.319 --> 00:22:10.160 say you're working with a data set of survey responses 501 00:22:10.839 --> 00:22:13.319 and some people just skipped certain questions. 502 00:22:13.039 --> 00:22:14.880 Right, So there are gaps in the data. 503 00:22:14.839 --> 00:22:17.519 Exactly, and those gaps can really skew your analysis if 504 00:22:17.519 --> 00:22:19.400 you're not careful. So you got to figure out how 505 00:22:19.440 --> 00:22:21.759 to deal with them. Sometimes you can just remove those rows. 506 00:22:22.079 --> 00:22:24.480 Other times you might replace them with a default value 507 00:22:24.920 --> 00:22:27.920 or use some statistical techniques to fill in the missing data. 508 00:22:27.960 --> 00:22:29.759 So it's not a one size fits all approach. You 509 00:22:29.799 --> 00:22:31.519 got to be strategic about it, right. 510 00:22:31.599 --> 00:22:34.920 Data cleaning is all about being thoughtful and understanding your 511 00:22:35.000 --> 00:22:37.799 data and what you're trying to achieve with your analysis 512 00:22:38.079 --> 00:22:38.599 makes sense. 513 00:22:39.599 --> 00:22:42.480 So we've talked about cleaning and preparing the data, but 514 00:22:42.640 --> 00:22:45.440 what about actually making sense of it. I mean, we 515 00:22:45.480 --> 00:22:47.960 can look at rows and columns of numbers all day, 516 00:22:48.079 --> 00:22:50.599 but that doesn't necessarily tell us anything useful. 517 00:22:50.839 --> 00:22:52.519 That's where data visualization comes in. 518 00:22:52.680 --> 00:22:56.599 Data visualization huh, so pretty graphs and charts. 519 00:22:56.279 --> 00:22:58.880 Well, it's more than just making things look pretty. It's 520 00:22:58.880 --> 00:23:02.599 about transforming that raw data into a visual format that's 521 00:23:02.640 --> 00:23:04.200 easy to understand and interpret. 522 00:23:04.319 --> 00:23:06.440 Okay, So, like, instead of just seeing a bunch of numbers, 523 00:23:06.519 --> 00:23:09.240 you can actually see patterns and trends exactly. 524 00:23:09.799 --> 00:23:13.559 A good visualization can help you identify trends, spot outliers, 525 00:23:13.880 --> 00:23:16.920 and see relationships between variables that might not be obvious 526 00:23:17.000 --> 00:23:18.359 just from looking at the raw data. 527 00:23:18.400 --> 00:23:20.799 So it's about bringing the data to life, making it 528 00:23:20.839 --> 00:23:22.400 more engaging, more intuitive. 529 00:23:22.680 --> 00:23:24.759 Right, it helps you tell a story with your data, 530 00:23:24.920 --> 00:23:26.119 you know, make it more impactful. 531 00:23:26.440 --> 00:23:29.799 Okay, I'm sold on the power of visualization. But what 532 00:23:30.039 --> 00:23:33.200 makes a good visualization versus a bad one? What should 533 00:23:33.240 --> 00:23:35.160 we keep in mind when creating them. 534 00:23:35.519 --> 00:23:40.079 A good visualization should be clear, concise, and informative. It 535 00:23:40.119 --> 00:23:44.039 should accurately represent the data without distorting or manipulating it 536 00:23:44.079 --> 00:23:46.640 in any way, and it should be easy to understand 537 00:23:46.720 --> 00:23:48.839 even for someone who's not familiar with the data. 538 00:23:48.920 --> 00:23:51.319 So no misleading graphs or anything like that. Got it. 539 00:23:51.400 --> 00:23:54.559 But there are so many different types of visualizations out there. 540 00:23:54.599 --> 00:23:56.039 How do you know which one to use? 541 00:23:56.359 --> 00:23:58.519 It depends on the type of data you have and 542 00:23:58.559 --> 00:24:01.200 what you're trying to show. Bar charts and line graphs 543 00:24:01.240 --> 00:24:04.400 are great for showing trends over time, scatterplots are good 544 00:24:04.400 --> 00:24:08.160 for exploring relationships between variables, pie charts are useful for 545 00:24:08.160 --> 00:24:11.119 showing proportions, and heat maps can be great for displaying 546 00:24:11.119 --> 00:24:13.799 more complex data in a visually intuitive way. 547 00:24:14.039 --> 00:24:16.200 Okay, so it's all about choosing the right tool for 548 00:24:16.279 --> 00:24:19.160 the job. But what about the audience? Do you have 549 00:24:19.240 --> 00:24:22.039 to consider who you're creating the visualization for. 550 00:24:22.440 --> 00:24:25.160 Absolutely, you need to think about who's going to be 551 00:24:25.160 --> 00:24:28.000 looking at this visualization and what they need to get 552 00:24:28.039 --> 00:24:30.559 out of it, What will resonate with them, what will 553 00:24:30.599 --> 00:24:31.880 help them understand the data. 554 00:24:32.359 --> 00:24:34.720 So data visualization is a bit of an art form, 555 00:24:34.759 --> 00:24:35.359 then huh? 556 00:24:35.400 --> 00:24:40.599 You could say that it requires a blend of technical skills, creativity, 557 00:24:40.839 --> 00:24:43.039 and a good understanding of communication. Principles. 558 00:24:43.240 --> 00:24:45.400 Okay, so now that I'm convinced of the importance of 559 00:24:45.480 --> 00:24:48.240 data visualization, what are some of the tools that are 560 00:24:48.279 --> 00:24:51.960 used to create these visualizations? Are there any specific Python 561 00:24:52.079 --> 00:24:53.480 libraries that are good for this? 562 00:24:54.160 --> 00:25:00.000 Absolutely, Python has several powerful libraries for creating amazing visualizations. 563 00:25:00.599 --> 00:25:02.839 Is one of the most popular and versatile that gives 564 00:25:02.880 --> 00:25:06.200 you tons of options. Seaborn is another one. It builds 565 00:25:06.200 --> 00:25:09.759 on that plotlib and provides a higher level interface and 566 00:25:09.839 --> 00:25:12.440 some pre built themes that make it really easy to 567 00:25:12.480 --> 00:25:15.200 create beautiful, professional looking visualizations. 568 00:25:15.279 --> 00:25:18.079 Okay, so matt plotlib is like the foundation and Seaborn 569 00:25:18.160 --> 00:25:20.519 is like adding those finishing touches making it all look 570 00:25:20.599 --> 00:25:21.480 polished and pretty. 571 00:25:21.640 --> 00:25:23.559 Yeah, that's a good way to think about it. And 572 00:25:23.599 --> 00:25:26.720 then there are other libraries for creating interactive visualizations, three 573 00:25:26.799 --> 00:25:30.279 D plots, even animated charts. It's really incredible what you 574 00:25:30.319 --> 00:25:31.680 could do with Python these days. 575 00:25:31.799 --> 00:25:33.880 It sounds like it. Okay, so we've talked about the 576 00:25:33.880 --> 00:25:36.480 power of data visualization, but let's shift gears for a 577 00:25:36.480 --> 00:25:39.839 moment and talk about bias in data analysis. I know 578 00:25:39.920 --> 00:25:42.680 we've mentioned ethical considerations before, but I'd like to dig 579 00:25:42.680 --> 00:25:46.039 into this a little bit more. How can bias sneak 580 00:25:46.119 --> 00:25:49.039 into data analysis? And what can we do to prevent it? 581 00:25:50.039 --> 00:25:53.359 That's a super important question. Bias is a huge issue 582 00:25:53.359 --> 00:25:55.079 in data analysis, and it can show up in many 583 00:25:55.079 --> 00:25:58.920 different ways. Sometimes the data itself is biased, reflecting existing 584 00:25:58.960 --> 00:26:03.039 societal prejudices or inequalities. Other times, the bias is introduced 585 00:26:03.119 --> 00:26:06.759 during the data collection, the processing, or the analysis itself. 586 00:26:07.079 --> 00:26:09.279 So the bias can either be built into the data 587 00:26:09.319 --> 00:26:12.400 from the start, or we can accidentally introduce it ourselves. 588 00:26:12.440 --> 00:26:14.440 That's a little scary, It definitely is. 589 00:26:14.680 --> 00:26:17.519 It's something we need to be constantly aware of. Let's say, 590 00:26:17.559 --> 00:26:21.319 for example, you're analyzing data on hiring practices and you 591 00:26:21.359 --> 00:26:23.519 find that men are being hired at a much higher 592 00:26:23.599 --> 00:26:27.559 rate than women. That discrepancy, well, it could be due 593 00:26:27.599 --> 00:26:30.720 to actual gender discrimination, but it could also be because 594 00:26:30.720 --> 00:26:35.000 of other factors like differences in qualifications or experience. It's 595 00:26:35.039 --> 00:26:38.640 really important to consider all the possible explanations and not 596 00:26:38.759 --> 00:26:39.799 jump to conclusions. 597 00:26:40.000 --> 00:26:42.960 So we need to be critical thinkers, questioning our assumptions, 598 00:26:43.000 --> 00:26:45.039 even when the data seems to point in a certain 599 00:26:45.079 --> 00:26:46.079 direction exactly. 600 00:26:46.119 --> 00:26:48.559 We have to be aware of our own biases, and 601 00:26:48.599 --> 00:26:50.519 we have to be aware of the biases that might 602 00:26:50.559 --> 00:26:53.400 already be embedded in the data, and we should always 603 00:26:53.400 --> 00:26:56.400 try to use data from different sources, involve people from 604 00:26:56.400 --> 00:26:59.799 different backgrounds in the analysis process. This can help mitigate 605 00:26:59.799 --> 00:27:00.720 potential bias. 606 00:27:01.119 --> 00:27:03.519 Makes sense. It's all about bringing a more diverse and 607 00:27:03.559 --> 00:27:07.160 inclusive perspective to the table. Speaking of perspectives, let's talk 608 00:27:07.160 --> 00:27:10.240 about storytelling again. You said earlier that the best data 609 00:27:10.279 --> 00:27:13.519 analysts are also good storytellers. Can you unpack that a 610 00:27:13.519 --> 00:27:16.079 little more for me? What does it actually mean to 611 00:27:16.119 --> 00:27:17.519 tell a story with data? 612 00:27:17.720 --> 00:27:21.000 It means going beyond just presenting the numbers and stats. 613 00:27:21.759 --> 00:27:25.400 It's about weaving those numbers into a narrative that grabs 614 00:27:25.440 --> 00:27:28.799 your audience's attention, helps them connect with the insights, and 615 00:27:29.000 --> 00:27:31.160 inspires them to maybe even take action. 616 00:27:31.440 --> 00:27:33.759 So you're not just presenting the facts. You're creating an 617 00:27:33.799 --> 00:27:35.279 experience for the audience. 618 00:27:35.839 --> 00:27:39.480 Exactly. A good data story should have a beginning, a middle, 619 00:27:39.480 --> 00:27:43.599 and an end. It should have characters, conflict resolution. It 620 00:27:43.640 --> 00:27:46.440 should make people feel something, make them want to do something. 621 00:27:46.680 --> 00:27:49.519 Wow. So it's about turning data into a horm of art. 622 00:27:49.759 --> 00:27:52.039 You could say that it's about taking something that can 623 00:27:52.079 --> 00:27:55.480 be dry and technical and infusing it with human emotion, 624 00:27:55.680 --> 00:27:57.279 with meaning, with purpose. 625 00:27:57.599 --> 00:28:00.200 That's a really powerful way to think about it. It 626 00:28:00.200 --> 00:28:03.359 makes me appreciate the potential of data analysis even more. 627 00:28:03.480 --> 00:28:05.599 It's pretty amazing, right, And we're seeing more and more 628 00:28:05.599 --> 00:28:08.720 examples of data storytelling and all sorts of fields, from 629 00:28:09.119 --> 00:28:12.839 journalism to science to even personal narratives. It's really becoming 630 00:28:12.839 --> 00:28:14.720 a pervasive part of our culture. 631 00:28:14.920 --> 00:28:17.480 It's like data is giving us this new language, this 632 00:28:17.559 --> 00:28:20.000 new way to connect with each other and understand the 633 00:28:20.039 --> 00:28:21.519 world around us exactly. 634 00:28:21.799 --> 00:28:25.240 And the more we embrace this language, the more powerful 635 00:28:25.359 --> 00:28:27.240 and impactful our stories will become. 636 00:28:27.519 --> 00:28:30.799 So what you're saying is anyone can be a data storyteller. 637 00:28:30.960 --> 00:28:34.640 Absolutely, It's not just for tech experts or statisticians. If 638 00:28:34.680 --> 00:28:36.960 you have a story to tell and you're willing to 639 00:28:37.039 --> 00:28:39.960 learn the tools, you can use data to make your 640 00:28:39.960 --> 00:28:43.839 story more compelling, more persuasive, and more impactful. 641 00:28:44.319 --> 00:28:48.519 Okay, I am officially inspired. Data analysis is more than 642 00:28:48.599 --> 00:28:51.000 just a skill. It's a way to make a difference 643 00:28:51.000 --> 00:28:51.359 in the world. 644 00:28:51.480 --> 00:28:52.279 That's exactly it. 645 00:28:52.359 --> 00:28:54.559 So before we wrap up this part of our deep dive, 646 00:28:54.599 --> 00:28:56.279 I just want to touch on one last thing that 647 00:28:56.319 --> 00:28:58.799 has really stuck with me throughout this conversation was that 648 00:28:59.079 --> 00:29:01.160 the importance of human curiosity. 649 00:29:01.279 --> 00:29:04.160 Ah. Yes, that's really the foundation of it all. 650 00:29:04.240 --> 00:29:07.279 It's what drives us to ask those questions, to explore 651 00:29:07.359 --> 00:29:09.599 new ideas, to seek out knowledge. 652 00:29:09.759 --> 00:29:13.960 Curiosity is the engine that fuels the entire data analysis process. 653 00:29:14.559 --> 00:29:16.480 Without it, it would just be a bunch of numbers 654 00:29:16.480 --> 00:29:19.119 in formula, is devoid of any real meaning or purpose. 655 00:29:19.240 --> 00:29:22.200 I love that curiosity is what transforms data into something 656 00:29:22.240 --> 00:29:24.079 truly meaningful exactly. 657 00:29:24.200 --> 00:29:27.359 It's that spark that ignites the fire, that fuels the 658 00:29:27.440 --> 00:29:31.119 passion for discovery, and that's what makes data analysis such 659 00:29:31.119 --> 00:29:32.839 a rewarding and exciting field. 660 00:29:33.160 --> 00:29:35.359 Okay, I think we've covered a lot of ground in 661 00:29:35.400 --> 00:29:37.839 this part of our deep dive. We've talked about the 662 00:29:37.920 --> 00:29:41.759 human side of data analysis, the importance of empathy, the 663 00:29:41.799 --> 00:29:45.559 power of storytelling, the nuances of data cleaning and preparation, 664 00:29:46.039 --> 00:29:49.599 and of course, the enduring importance of human curiosity. But 665 00:29:49.680 --> 00:29:51.400 before we move on to the final part, I think 666 00:29:51.400 --> 00:29:55.480 it's also important to acknowledge that data analysis isn't perfect, right. 667 00:29:55.279 --> 00:29:57.279 It's important to be aware of its limitations. 668 00:29:57.519 --> 00:29:58.799 What are some of the things we need to be 669 00:29:58.839 --> 00:29:59.559 cautious about. 670 00:29:59.759 --> 00:30:03.240 Well, for one, data can only tell us about the past. 671 00:30:03.599 --> 00:30:06.480 It can't predict the future with absolute certainty. 672 00:30:06.240 --> 00:30:08.039 So we can't treat it like a crystal ball. 673 00:30:08.480 --> 00:30:12.000 Exactly. We can use data to identify trends and make 674 00:30:12.119 --> 00:30:15.279 educated guesses about what might happen, but we should always 675 00:30:15.279 --> 00:30:18.519 be careful about extrapolating too far beyond the data we have. 676 00:30:18.759 --> 00:30:21.799 Okay, so don't get too carried away with predictions. What else? 677 00:30:22.119 --> 00:30:26.000 Another limitation is that data is often incomplete or imperfect. 678 00:30:26.440 --> 00:30:29.400 There might be gaps in the data, errors in data entry, 679 00:30:29.799 --> 00:30:32.440 or biases in the way the data was collected. 680 00:30:32.400 --> 00:30:34.599 So we can't just blindly trust the data. We need 681 00:30:34.640 --> 00:30:36.440 to be skeptical and critical thinkers. 682 00:30:36.559 --> 00:30:38.960 Exactly, you always have to question the data, where it 683 00:30:39.000 --> 00:30:42.000 came from, how it was collected, and what limitations it 684 00:30:42.119 --> 00:30:46.000 might have. And finally, it's crucial to remember that data 685 00:30:46.079 --> 00:30:49.000 is just one piece of the puzzle. We should always 686 00:30:49.039 --> 00:30:54.519 consider other sources of information like qualitative research, expert opinions, 687 00:30:54.799 --> 00:30:58.319 and even our own intuition and experience. Data can be 688 00:30:58.359 --> 00:31:01.599 a powerful tool for informing our decisions, but it shouldn't 689 00:31:01.599 --> 00:31:02.480 be the only tool. 690 00:31:02.640 --> 00:31:05.359 That's a really good point. We can't let data analysis 691 00:31:05.440 --> 00:31:08.480 become a substitute for good judgment and critical thinking. It's 692 00:31:08.519 --> 00:31:11.920 about using data to enhance our understanding, not to replace it. 693 00:31:12.039 --> 00:31:15.559 Absolutely, data analysis is a tool, and like any tool, 694 00:31:15.599 --> 00:31:17.799 it needs to be used wisely and responsibly. 695 00:31:18.079 --> 00:31:20.440 So with those limitations in mind, I'm curious to hear 696 00:31:20.480 --> 00:31:22.960 your thoughts on the future of data analysis. Where do 697 00:31:23.000 --> 00:31:24.279 you see this field heading? 698 00:31:24.559 --> 00:31:27.880 That's a tough question. It's such a rapidly evolving field. 699 00:31:28.400 --> 00:31:30.400 It's hard to say for sure what the future holds, 700 00:31:31.039 --> 00:31:34.039 but there are definitely a few trends I'm excited about. Okay, 701 00:31:34.240 --> 00:31:37.039 Like what One trend that's already having a huge impact 702 00:31:37.359 --> 00:31:41.000 is the rise of artificial intelligence and machine learning. I 703 00:31:41.000 --> 00:31:43.599 think we're just scratching the surface of what's possible with 704 00:31:43.640 --> 00:31:44.559 these technologies. 705 00:31:44.759 --> 00:31:47.759 So AI is more than just a buzzword. It's really 706 00:31:47.839 --> 00:31:50.319 changing the game when it comes to data analysis. 707 00:31:50.359 --> 00:31:54.000 Absolutely, AI can help us analyze these massive data sets, 708 00:31:54.440 --> 00:31:57.759 identify patterns that we might not even see, make predictions 709 00:31:57.759 --> 00:32:01.519 with much greater accuracy than before. It's pretty mind blowing. 710 00:32:01.759 --> 00:32:04.079 But we have to remember AI is a tool, and 711 00:32:04.240 --> 00:32:06.599 like any tool, it can be used for good or. 712 00:32:06.559 --> 00:32:09.680 Bad, So it's important to think about the ethical implications 713 00:32:09.720 --> 00:32:12.680 of AI, especially as it becomes more integrated into data 714 00:32:12.680 --> 00:32:13.839 analysis exactly. 715 00:32:13.920 --> 00:32:16.720 Another trend I'm really excited about is the growing emphasis 716 00:32:16.720 --> 00:32:17.720 on data literacy. 717 00:32:17.839 --> 00:32:20.480 Data literacy, so you mean like everyone needs to become 718 00:32:20.480 --> 00:32:21.359 a data expert. 719 00:32:21.599 --> 00:32:25.400 Well, not necessarily an expert, but as data becomes more 720 00:32:25.440 --> 00:32:28.079 and more a part of our lives, it's crucial that 721 00:32:28.119 --> 00:32:32.119 everyone has at least a basic understanding of how to 722 00:32:32.160 --> 00:32:33.920 interpret and critically evaluate it. 723 00:32:34.160 --> 00:32:36.519 Okay, so we need to be able to spot misinformation, 724 00:32:36.839 --> 00:32:39.599 to understand what the data is really telling us exactly. 725 00:32:39.640 --> 00:32:42.920 That's data literacy. It's about being able to think critically 726 00:32:42.920 --> 00:32:46.319 about data and make informed decisions based on that data. 727 00:32:47.000 --> 00:32:48.759 And lastly, I think there's going to be a huge 728 00:32:48.759 --> 00:32:51.960 demand for data storytellers in the future, people who can 729 00:32:52.039 --> 00:32:55.200 bridge that gap between the technical world of data and 730 00:32:55.240 --> 00:32:56.839 the human need for meaning and connection. 731 00:32:57.079 --> 00:32:59.240 It's not enough to just crunch the numbers. You need 732 00:32:59.279 --> 00:33:01.359 to be able to commit unicate those insights in a 733 00:33:01.359 --> 00:33:02.640 way that resonates with people. 734 00:33:02.799 --> 00:33:06.039 You got it. Data storytelling is becoming an essential skill 735 00:33:06.039 --> 00:33:07.240 in pretty much every field. 736 00:33:07.440 --> 00:33:10.480 Wow, these are some exciting trends. So to wrap up 737 00:33:10.480 --> 00:33:12.240 this part of our deep dive, I guess what you're 738 00:33:12.279 --> 00:33:15.359 saying is that the future of data analysis is bright, 739 00:33:15.799 --> 00:33:18.519 but it also comes with a lot of responsibility. We 740 00:33:18.599 --> 00:33:21.400 need to be mindful of the ethical considerations, we need 741 00:33:21.400 --> 00:33:24.519 to promote data literacy, and we need to never lose 742 00:33:24.559 --> 00:33:27.440 sight of the human element, the power of storytelling. 743 00:33:27.640 --> 00:33:31.440 Well said, data analysis is a powerful tool, but it's 744 00:33:31.519 --> 00:33:35.160 ultimately up to us as humans to use it wisely 745 00:33:35.319 --> 00:33:36.640 to make the world a better place. 746 00:33:37.000 --> 00:33:38.839 I love that it's not just about the data, it's 747 00:33:38.880 --> 00:33:41.000 about what we do with it, how we use it 748 00:33:41.039 --> 00:33:43.880 to make a positive impact. Okay, I think that's a 749 00:33:43.880 --> 00:33:45.920 great place to pause for now. This has been an 750 00:33:45.920 --> 00:33:47.920 incredibly insightful conversation. 751 00:33:48.200 --> 00:33:50.480 I agree, we've covered so much ground, and I'm really 752 00:33:50.519 --> 00:33:53.200 looking forward to continuing our exploration in the final part 753 00:33:53.240 --> 00:33:53.880 of our deep dive. 754 00:33:54.000 --> 00:33:55.599 Okay, so we're back for the final part of our 755 00:33:55.680 --> 00:33:58.200 data analysis deep dive. We've talked about so much already, 756 00:33:58.200 --> 00:34:00.960 from the technical stuff like Python and webster graping, to 757 00:34:01.160 --> 00:34:04.079 the ethical side of things and the art of data storytelling. 758 00:34:04.319 --> 00:34:05.599 What else is there to explore? 759 00:34:05.799 --> 00:34:08.360 Well, you know, we talk about data as this tool 760 00:34:08.480 --> 00:34:13.119 for uncovering truths and telling stories, but we shouldn't forget 761 00:34:13.199 --> 00:34:16.320 about its potential for actually solving problems, making a real 762 00:34:16.360 --> 00:34:17.119 difference in the world. 763 00:34:17.199 --> 00:34:19.239 That's a great point. I think we often get caught 764 00:34:19.280 --> 00:34:21.559 up in the analytical side of things, you know, just 765 00:34:21.599 --> 00:34:24.079 trying to understand the data. But you're right, it can 766 00:34:24.119 --> 00:34:27.880 be used to address real world issues and actually create change. 767 00:34:28.039 --> 00:34:31.039 Absolutely. I mean, think about the big challenges facing our 768 00:34:31.039 --> 00:34:36.800 world today, climate change, poverty, disease. Data analysis is already 769 00:34:36.800 --> 00:34:40.880 being used to develop solutions, track progress, and hold people accountable. 770 00:34:40.960 --> 00:34:44.679 It's like data analysis is the bridge between information and action. 771 00:34:44.800 --> 00:34:47.159 We can use it to understand the problems and then 772 00:34:47.239 --> 00:34:49.480 actually do something about them exactly. 773 00:34:49.920 --> 00:34:53.159 It's pretty amazing what's happening in so many fields. Organizations 774 00:34:53.199 --> 00:34:56.719 are using data analysis to optimize energy consumption, reduce ways, 775 00:34:56.800 --> 00:34:58.559 and develop more sustainable practices. 776 00:34:58.760 --> 00:35:01.079 So data can help us tackle climate change head on. 777 00:35:01.199 --> 00:35:02.760 What about other areas well? 778 00:35:02.760 --> 00:35:06.679 In healthcare, data analysis is being used to identify disease 779 00:35:06.719 --> 00:35:11.079 outbreaks early on, track the effectiveness of treatments, and even 780 00:35:11.159 --> 00:35:12.360 personalized medicine. 781 00:35:12.360 --> 00:35:15.599 Wow, that's incredible. It's really inspiring to see how data 782 00:35:15.599 --> 00:35:18.119 analysis is being used to make a tangible impact in 783 00:35:18.159 --> 00:35:20.800 the world. But I'm also curious about the future of 784 00:35:20.920 --> 00:35:24.719 data analysis itself. What skills and knowledge do you think 785 00:35:24.719 --> 00:35:27.920 will be most valuable in this field as it keeps evolving. 786 00:35:28.079 --> 00:35:30.480 Hmm, that's a tough one to predict. I mean, the 787 00:35:30.519 --> 00:35:33.480 field is constantly changing. Technical skills are always going to 788 00:35:33.480 --> 00:35:37.119 be important, obviously, but I think those critical thinking skills, 789 00:35:37.199 --> 00:35:40.639 problem solving and communication, those are going to become even 790 00:35:40.639 --> 00:35:41.360 more valuable. 791 00:35:41.679 --> 00:35:43.559 Right. It's not enough to just crunch the numbers. You've 792 00:35:43.519 --> 00:35:45.320 got to be able to explain what they mean, what 793 00:35:45.360 --> 00:35:47.239 the implications are exactly. 794 00:35:47.559 --> 00:35:49.639 Data analysts of the future, they need to be able 795 00:35:49.719 --> 00:35:53.000 to not only analyze data, but also interpret it, put 796 00:35:53.039 --> 00:35:55.679 it in context, and then communicate those findings in a 797 00:35:55.679 --> 00:35:57.920 way that's clear, compelling, and actionable. 798 00:35:58.199 --> 00:36:00.639 So it's about being a well rounded thinker, not just 799 00:36:00.679 --> 00:36:02.639 a technical whiz exactly. 800 00:36:02.960 --> 00:36:05.199 And I also think we're going to see a greater 801 00:36:05.280 --> 00:36:08.760 need for data analysts who really understand ethics, who take 802 00:36:08.840 --> 00:36:12.719 social responsibility seriously. As data becomes more and more powerful, 803 00:36:12.840 --> 00:36:14.760 we need to make sure it is being used ethically 804 00:36:14.800 --> 00:36:15.760 for the good of everyone. 805 00:36:15.840 --> 00:36:18.840 Absolutely, data can be a powerful tool for good, but 806 00:36:18.920 --> 00:36:21.960 it can also be misused. We need to make sure 807 00:36:22.079 --> 00:36:25.679 being used to create a more just and equitable world, 808 00:36:25.880 --> 00:36:28.280 not to reinforce existing inequalities. 809 00:36:28.320 --> 00:36:30.440 I couldn't agree more. And on that note, I think 810 00:36:30.480 --> 00:36:32.760 it's fitting that we circle back to something you mentioned earlier, 811 00:36:33.320 --> 00:36:35.320 the importance of human curiosity. 812 00:36:35.599 --> 00:36:39.440 Yes, curiosity is what has driven this whole conversation. Really, 813 00:36:39.719 --> 00:36:43.039 it's what drives us to explore and learn and understand. 814 00:36:43.199 --> 00:36:46.760 Curiosity is the fuel that powers the engine of discovery. 815 00:36:46.800 --> 00:36:50.519 Without it, data analysis would just be a dry, technical exercise. 816 00:36:50.559 --> 00:36:52.559 It wouldn't have that spark, that sense of wonder. 817 00:36:52.679 --> 00:36:55.360 It's like curiosity is the secret ingredient that makes data 818 00:36:55.400 --> 00:36:57.719 analysis so engaging, so rewarding. 819 00:36:58.039 --> 00:37:00.599 Exactly. So, my final thought for one out there who's 820 00:37:00.639 --> 00:37:04.239 interested in data analysis is this, never lose that sense 821 00:37:04.239 --> 00:37:09.079 of wonder, never stop asking questions, and never be afraid 822 00:37:09.119 --> 00:37:10.519 to challenge the status quo. 823 00:37:10.880 --> 00:37:13.199 I love that, and I think it's the perfect message 824 00:37:13.199 --> 00:37:15.400 to end on this whole deep dive has been an 825 00:37:15.400 --> 00:37:17.880 incredible journey. It's open my eyes to the power of 826 00:37:17.960 --> 00:37:22.800 data analysis, the complexities, the challenges, but also the immense possibilities. 827 00:37:23.119 --> 00:37:26.400 I'm walking away feeling inspired and ready to dive even 828 00:37:26.440 --> 00:37:28.000 deeper into this fascinating world. 829 00:37:28.239 --> 00:37:30.440 It's been a pleasure exploring these ideas with you. I 830 00:37:30.440 --> 00:37:32.360 hope you all out there listening feel the same way. 831 00:37:32.559 --> 00:37:34.440 Thank you so much for joining us on this deep 832 00:37:34.480 --> 00:37:37.400 dive into the world of social media data analysis. We 833 00:37:37.440 --> 00:37:40.320 hope you've enjoyed the ride, learn something new, and maybe 834 00:37:40.440 --> 00:37:42.920 even sparked your own curiosity about the power of data. 835 00:37:43.800 --> 00:37:46.360 And remember, this is just the beginning. The world of 836 00:37:46.440 --> 00:37:49.039 data is vast and ever changing, and there are endless 837 00:37:49.079 --> 00:37:52.760 stories waiting to be uncovered. So keep exploring, keep learning, 838 00:37:52.880 --> 00:37:57.639 and keep asking those questions. Until next time, happy analyzing.