 Hello out there to everybody. I'm going to give everyone a second to let their audio kick in and get fully logged on before we get going. I'm excited to see all of these participants looking to learn more about data science in the School of Information Studies at Syracuse University. My name is Ian Richardson. I'm the undergraduate recruitment specialist here at the School of Information Studies or the iSchool at Syracuse University. And again, we're super excited to have you join us. I feel like we're going to have a very interesting and exciting class who's going to be taught by one of our assistant professors, Jezmina Takeva. And so without further ado, I'm going to pass it over to Professor Takeva and she's going to talk about what Snapchat and Walmart have in common and the answer of course is data science. So Jezmina, I'm going to make you the host so that you can share your screen. And sounds good. It's all yours. Thank you. All right, everyone. So maybe before. Oh, does that mean that I need to let people in? No, I can. All right. All right. I wasn't sure if you have access and waiting room. All right, perfect. Okay, so I'll try to keep an eye on the chat as I'm presenting, but I'd like to welcome everybody. I know it's been a long day and maybe before we get started, if you'd like to just say where you're joining us from, you could write that in the chat as I'm getting ready to start the slideshow here. Oh, great. Illinois, Hong Kong. Oh, wow. Virginia. Perfect. Well, I'm originally from Bulgaria. I don't know if you know where my country is, but I did my undergrad in the United States and then I did my PhD and now I teach at the iSchool. So what we're going to be talking about today is what is something that Snapchat and Walmart have in common? And as Ian said earlier, one of the answers is data science. So what we have prepared for today is the things that you're going to learn after we're done with today's lecture. We're going to be able to answer questions such as what is data? What is data science? Who does data science and what do these people do exactly and also why it matters? But before we go there, I'd like to invite everyone to just take your phone and point your camera to this QR code on the slide. I hope that everyone can see it. And then it's going to take you to a poll that has one single question. So all that's required of you is to just enter an answer. It can be one word and if anyone has already done this, you can see that the question is what your favorite app is. So I'm really curious to see what everyone enjoys using. And I'll give us maybe a minute and then we can take a look at the answers. All right. So let me just see if I can, all right. So can everyone see the word cloud that's starting to form on my screen? Someone can just give me a thumbs up for some other reaction. Okay, perfect. So for those of you who are looking at the screen right now, and again, we're going to give ourselves a couple more seconds. We have 28 votes so far. Folks should feel free to keep voting. But who seems to be the winner so far? Can someone help me decipher this word cloud? Because this is, if you think about it, a source of data, right? So how are we to interpret this data that we see in front of us? What are some of the, yeah, exactly. So we see right in the center, size here, the font size sort of corresponds to the number of votes that each of these apps is receiving. So we can see that the leader so far is Instagram. Yep, that's exactly right. And Snapchat and QuickDoc I guess are 12 seconds. All right. So that's very exciting. I can see a lot of Instagram users on here. And that's why in a way we chose to call this lecture Snapchat and Walmart, because we know how important these apps are as part of our day-to-day experience on one hand. And then the next question that I have for you has something to do with this word cloud. So if I go back to my slides, I want to point your attention to this next slide over here, which I'm going to see if I can come back in presentation mode. All right. So we can all see the logos of some of the major companies on the market so far. We already saw the word cloud as well with apps such as Snapchat, QuickDoc, Instagram, Facebook, Twitter, and so on. So what I want us to do is take a couple of seconds again and just think about some things that all of these companies may have in common. Does anything come to mind? And again, feel free to write your answer in the chat. When you think about all of these companies, they come from different industries, right? So industry is not the common denominator here. Yep, that's exactly so. Jack gave us the correct answer. They all operate with lots of data. Now, whether this is data that they themselves generate, or that their customers generate, and that they have access to, or maybe data that they acquire externally from other sources. But the point is that all of these companies operate with really tons and tons of data. So why is it so important then to talk about data? Here's another question for you. And again, just feel free to point your phone camera to the QR code, and then it will take you to a question that we're going to open up and look at in just a few seconds. But I just want to give everyone the chance to open it on their phone. All right. Maybe a couple more seconds. And then I'll switch over or rather I'll keep it on my screen just so people can still have access to the question. But at the same time, I want to try and access the answers to see can let me just monitor the chat. Everyone has access to this next question, right? And the question is, what comes to mind when we think of the word data? Now, let's take a brief look at your answers. And again, I apologize. It's a little bit hard for me to access the right tab here with so many top up windows. But I think I'm almost there. Yeah. All right. All right. So people are saying what comes to mind are things like information, numbers, statistics. Okay, great. Yeah. So these are pretty much related concepts. And coming back to the slides, let's see why these things are important. How come someone is saying things like hackers too? There's the internet cloud, there's phones, there's Zuckerberg demographics. So in a way, we're going to be touching on each of these in a sort of a brief fashion tonight. But it's really cool that you guys thought of these connections because they really do exist and are terribly important. So coming back to the slides, let's see exactly how we can relate all of these concepts. Can we make sense of this word cloud? All right. Maybe one way of starting to answer this question is by first defining data. So if you were to think of a definition of data, the easiest way to think of data is data is any fact or value or observation that exists in the world. And that's very important to keep in mind because we're now in the internet age. There's a lot of sources of data that live online, but that does not exhaust the full spectrum of data sources. But focusing on online data for this lecture. Here I have a question for you, and I'll take a look at the chat again because this is where I'll invite you again to write your answer. And I'll go one by one through this list, and I'm going to be asking you the following. Do you think that each of these items constitutes a data item? Starting with A, a numeric cell in an Excel spreadsheet. Do you think that this represents data? We think of this as data. Yup, that's right. So this is in a way one data item. Now how about B? Is it text message data? Yup, that's right. B is also part of what we think of when we talk about data. Now how about an Instagram post? All right. I see an overwhelming stream of yeses, and you are correct again. Instagram posts can also be considered data. But how about a YouTube video? Can we think of a YouTube video as a form of data? Yup, the answer is yes again. Now let me ask you the opposite question. Are any of the things on this list not data? Is there anything on this list that you see which is not a form of data? Yup, that's correct again. So every single thing that you see on this list in a way represents data. Why? Because it fits the description that we gave up here. So it does represent either a fact or a value or an observation about the world, including this last one, which is simply something like on November 10th at 7 p.m., person X bought a carton of eggs from a Walmart or something like that. Even this in itself represents data because it can be aggregated with other data points, and it can be turned into some useful insights. All right. So let's take a look now, knowing and sort of keeping in mind the definition of data. This next question, which is asking us if the things that we see, and again, it's the exact same list of items, is each of them structured or not? Now, can someone help me out in the chat and try and give us an idea about what we could possibly mean by structure? What does structured data look like? Can anyone give me an example? When we say that something is structured, what do we mean? And specifically in the context of data, does anyone have any sort of guesses? Okay. So Oscar says a graph, something organized and recorded. Yep. That's a very good explanation, Caitlin. Yep. Something organized, formatted. Yep. So I'll accept all of these. When we talk about structured data, we mean data which is ready to be analyzed. Okay. So you can think of structured data as data which is ready, ready to be operationalized. It's ready to be sort of the subject of mathematical operations. For example, numerical column in Excel, we can apply a bunch of statistical functions to that column in order to get information out of it, such as what is the mean value or sort of the average, right? The mean and the average, that's the same thing. So we can talk about the average number of likes for Facebook posts and so on. So something that we can neatly store in an Excel spreadsheet and sort of apply mathematical operations to something that is ready to be analyzed, that's what structured data is. So with that definition in mind, let's go down the list and try and figure out if each of these items is structured or not. So if we think of a numeric cell, so a cell in an Excel spreadsheet that contains a number, can we think of this as structured data? Yep. Justin, that's a great answer. Yep. So this is correct. Number A is indeed an example of structured data. How about a text message? B. So here people are starting to feel a little less certain. So Rihanna says that it is structured, but Finley says no. So who's correct? And for those of you saying no, why do you think that is? Like, why can't we think of a text message as a form of structured data? What's unstructured about it? Any ideas? All right. So let me give you a hint. So a text, exactly. Yep, that's right. It's less organized. And it doesn't have to have data in it all the time. So if we think of data as numbers, yeah, so not all text messages contain numbers or values and other sort of numerical things like that. But we can still think of the message overall, the text message as a form of data, maybe not in and of itself, but let's say that we have access to a vast number of text messages over time. We can mine these, we can apply, for example, techniques known as text mining to figure out what the overarching topic of conversation is and so on. So while a text message is still very much data, it's not necessarily structured because we cannot take a bunch of text messages and readily apply some analytical tools to them. Like, we can't take the average of text, right? So it's not structured. Now, how about an Instagram post? Is an Instagram post? Yep, that's correct. Following the same logic, we can't readily take the average of an Instagram post because it's a photo, right? Most of the time. Or even if it's just text, the same logic applies as the one that we discussed with V, a text message. It's not ready for analysis just yet. It needs to be pre-processed before we can analyze it further. Now, how about a YouTube video? All right, so, Kailin, that's a great answer. Do companies only depend on structured data? And I believe that I have an answer for you in my next slide. So, but before we get there, yes, I do want to confirm that a YouTube video is also an example of unstructured data. So it is not structured for the same reason that an Instagram post and a text message cannot be considered structured data because they need further pre-processing. Now, how about E, a store transaction record? Yep, so now this is in tabular form, right? It's part, it sort of lives in a database already, which means that it has already been pre-processed. There have been some steps prior to it being entered into the database that sort of have made it more tractable and sort of ready to be operationalized and analyzed. So great job, everyone. So we tackled this question successfully. And now let's take a look at sort of a visual representation of what data looks like in the wild, so to speak. Like, what do we see when, what a data scientist sees when they look, for example, at a web page, such as this one. Does anyone know what website this is from? A website where instead of star ratings, you get these sort of egg icons? Yeah, so it's very close to Amazon. It's sort of similar. Yep, that's exactly why Noah gave us the correct answer. It's a website called Newegg, which is sort of kind of like Amazon, but it's geared specifically towards people looking to buy hardware. So it's mainly focused on computer products. But in every other aspect, it closely resembles Amazon. In that, it's a sort of platform where you can shop for these things. And just like Amazon, you can leave reviews and you can rate products. And that's exactly what we see here. We see sort of a product that has been rated five out of five, not stars but eggs. And then we even have stuff like, of course, a product name, how many questions people have asked about it, how many answers have been given to these questions, a brief description of the product as well. And then we have sort of a sample of reviews. We have one review that we see displayed here. And so my question now is, can someone help me figure out what in this image could constitute a structured, more or less structured form of data? And you see these sort of orange rectangles here, they just represent different elements, different data points. So can someone give an example of structured, like some part of this page that you believe represents structured data? Price, that's very good. Exactly. So price, it's a numerical variable already. So yes, we can readily analyze this ratings, that's another good one. Yep, percentages of reviews, that's, yep, so great. So I believe that you grasp that concept of structured versus unstructured data. But let me ask you this then, number of reviews, yeah, sample size, that's another great one. Time the review was uploaded, oh so very good. Now how about on the flip side, if we were to ask about unstructured data, what could be an example of that here? Something that needs further processing before we can analyze it. Okay, name, yeah, we can't really compare names in any numerical fashion, other than maybe looking at the length of the name or something like that, but we can't get an idea about the content of these variables without further processing. Content of the review, yep, description of the product, pros and cons, that's all great and correct answers, yep, to review the product description. Yep, so you got things that we see that contain text here and are not countable, they can be considered as unstructured data. All right, so I think you guys got this. Now let's take a look coming back to the question that was asked earlier. Do companies only rely on structured data? And the answer is overall the data market, which whose representation we see here, is showing us that the amount of data is growing exponentially over time. And there are different layers of data here, as we can see enterprise data or what we can think of as the best representative of structured data. Everything that's neatly stored in tabular form, in databases, in spreadsheets, that flows the clean organizations that can be readily analyzed, everything can sort of or the best representation of that can be found in this bottom layer here. So as we can see, this layer is not where we see the biggest growth overall. The next layer that's sort of not as prominent, but it's still growing, is does anyone know what the OIP stands for? And let me just access my chat to see has anyone heard of this abbreviation before the OIP? Yep, exactly voiceover IT. So what could be an example of that? One example of that is for example Skype, right? It's any sort of phone conversation that can be had online. Okay, so we can see that this sort of layer is also growing. And then another interesting one though is this green one, which represents social media and web data. So we can see substantial growth over there too. But by far, the layer that is growing at the highest rate, fastest rate is this top layer here, which represents sensor data and devices. Now we can think of these three layers on top here as sort of jointly representing unstructured data. This is data that is not necessarily always structured. In fact, this is sort of user content that comes to us in the form of unstructured elements like text, like images, like videos, like audio files and so on. So this is by far the biggest chunk of data growth that we observe. It happens in sort of the domain of unstructured data. So to answer the question about whether companies only rely on structured data, by far and increasingly companies have started relying on unstructured data as well. And now when we talk about sensor data, what we can say is that what can be measured on the internet and what can enter the realm of data is not anymore just what we do, the sort of websites that we browse, the buttons online that we click and so on. It's way deeper than that. Even things that we say can be recorded and stored as data. Does anyone know how this could happen with any type of device, a smart device that is under the umbrella of the internet of things? What are some devices that can record what we say and then store it in the form of data? Something, yep, so phones, the personal assistant kind of Alexa. Yep, that's exactly what I was looking for. So things like Siri, like Alexa, like Google Home, these devices come with sensors that can make use of even things like our speech, right? The things that we say can also be operationalized and stored in data. But it goes even deeper than that with these, for example, step trackers and other type of activity trackers like that, wearable technologies, even the number of steps that we take every day, even this can be represented as data, right? And even deeper than that, based on our activities online, companies, big tech companies, can assess things as immaterial as our very thoughts. So how we feel about things, what we think about, what we dream of, what our aspirations are, even these things can be accessed simply through our online behavior. All the tests that you take online, like personality quizzes and all that, the answers that you provide to those can be turned into data, but it goes even deeper than that. If anyone is familiar with the Cambridge Analytica scandal, what happened there is not necessarily that millions of American voters took personality quizzes online. It went deeper than that. Only a fraction of Facebook users did in fact take these personality quizzes and then, based on the data that developers got from these quizzes, they trained a machine learning model that could relate your psychological traits, your personality traits to your Facebook activities. Like, for example, they could say things like people that are more extroverted like pages on Facebook that have to do with sports and with outdoors activities. And so having this model in place, they didn't even need Facebook users to complete these online quizzes to establish what their personality traits were, but they could just look at the pages that Facebook users have liked on the platform in order to create sort of a psychological profile of each person. That's something that's known as psychometrics, and this is exactly how companies can access deep layers of our personality. All right. So here's another question for you. Who do you think knows you better? Facebook or your friends? And again, it's not a trick question. I just want to see what people think. We already have an answer, so people think that it's Facebook, even though some of you do believe that your friends know you better. Okay, to be perfectly honest with you, it might have been a little bit of a trick question. And how do I mean this? What exactly do I mean by know you? Right? Like, this is a legitimate question because it really depends. So the correct answer here depends on our definition of knowing somebody. What does it mean to say that you know someone? Is it enough to sort of just be able to predict their personality, like whether they're more outgoing or whether they're more introverted and so on? Like, is that enough to say that you know somebody or do we mean that you know their, let's say, people's secrets or you've known them for years since they were a child and so on? So I am referring here to the first more superficial definition of knowing someone, knowing in the sense of being able to predict their future behavior. And so with that definition in mind, in terms of data, correct. Yes, Amira, Facebook, in that case, would know us better. Now, how does that work? There's a very controversial social experiment that Facebook performed a few years ago. And Facebook developers were able to find out that it didn't really take all that much data from an individual Facebook user to be able to predict their behavior. And how this works is really scary, if you think about it, because they found, for example, that based on just 10 of your likes on Facebook, so just by having access to about 10 of the pages or groups that you have liked on Facebook, they could predict more accurately your behavior than a classmate of yours. And with something like 70 likes, this prediction could be more accurate than most of your friends. And when we increase the number of likes to about 300, then they were able to sort of predict your behavior better than even your romantic partner. So that just goes to show how powerful the sort of data breadcrumbs that we leave online with our online behavior, how they can be aggregated and analyzed in a way to create these powerful models that can really get to predict things about us in a very detailed and scarily accurate way. That's why many experts today say that the world's most valuable resource is no longer oil but data. And the reason data is so powerful is precisely because when companies have access to large amounts of data, now don't think of it just in terms of an individual user. My data in and of itself may not be very helpful to companies like Facebook, but the power of Facebook stems from the fact that they have access to data not just of me, but of millions of other users like me. And basically by combining all our data together, they're able by having access to all this data, they're able to create these powerful algorithms that can really tell us a lot about society. Now, does anyone know what this logo in the lower right hand corner here is? Have you seen this before? It looks like a warped infinity symbol. Meta, exactly. So what is meta? It's sort of the rebranded company behind Facebook now. Exactly. So Facebook does have access to really an unprecedented amount of our personal data. And we're going to be talking about how exactly this constitutes its power, but also its responsibility in a little bit. But before we go there, I guess the next question in order is who creates these powerful models that can predict your personality better than your family? Does anyone have any idea what we call these people that are behind these algorithms, behind these models? Data analysts or data scientists? Which one is it? And the answer to this. So it's a good question of what the difference is between a data analyst and a data scientist. Does anyone know? Yes. So the correct answer here is data scientists. But what is the difference between a data scientist and a data analyst? Okay. So let me show you the definition of a data scientist. So a data scientist is involved in all of these activities that you see on here. So they can code, they can pre-process data, they can visualize data, they can apply machine learning techniques to data, they can present their findings to other stakeholders, and they also know things about ethics, ideally, hopefully. Exactly. So Sam gave us a great answer. Analysts interpret the data, data scientists sort of either create or create the models for. So data analysts most of the time take data that already exists and they don't really participate in the creation of the algorithms that can analyze that data. That's more in the purview of data scientists themselves. So data scientists are people that are involved in, according to the technical definition of data science here, they have very particular and deep domain knowledge. They also know math and statistics. They are well versed in computer science and their good communicators. This is a technical definition of data science, but the practical definition by the person who coined the term data science, DJ Patel, who you see here. So he's the one that gets credited with introducing the idea of data science into the mainstream. And so he believes that data science can most broadly be defined as anything that uses data in novel ways to build things. So, as Sam told us in the chat, while data analysts analyze data and don't necessarily create things out of it, other than, you know, an analysis, when it comes to data scientists, they create algorithms that can further operationalize and sort of leverage that data. Okay, now let's talk briefly about some common misconceptions about data science. And this one, I hear a lot. And I started with this one because this is where people usually give up when no matter how much they are interested in data science, when they hear the word math or computer science, they sort of get scared and they think that data science is not for them. But I would like to tell you all that it's wrong that you must be a math or computer science was in order to be a good data scientist. And the reason why it's wrong is because data science and sort of the expertise that data scientists have looks more like a T. It's a T-shaped type of thing of the kind that you see here. And what I mean by that is, yes, you do need to be aware of certain data science principles. You do need to be aware of certain statistical principles. But really the core of it is your own domain sort of area of expertise, which does not necessarily have to be data science. It can be, for example, journalism. You may be really interested in journalism and you can be a data science journalist, by which I mean journalists that use data that they find online or that they compile themselves. And then they create powerful machine learning algorithms in order to analyze it and to write new stories about it. But again, really the depth here comes from their domain knowledge in journalism and not so much from math or computer science. Now, a second common misconception about data science is that it's just the fact it won't last long. We're sort of past the peak already. And I want to emphasize that this is wrong as well because there are many projections by market experts that say that in the next few years there's got to be a shortage of 250,000 data scientists in the United States alone. So just think about the scale of this. And the reason why there is or the market for data scientists keeps expanding is because of the graph that I showed you earlier, this exponential growth of data. Well, guess what? As data is growing, so does the need for people to analyze it and to create models to help that analysis, right? So we naturally see more and more need for data scientists in the current economy. And then the third misconception has to do with sort of the idea that artificial intelligence would somehow replace data scientists. So basically the algorithms will replace humans. Now, this is also wrong and it's wrong because of where we stand with regards to the singularity. Does anyone here know what the singularity is? Have you ever heard of this term before? The technological singularity. If you've seen any sci-fi movies like X, Mahina and so on, like the idea of the singularity is that moment in time when the machines will take over, where technological advances will be so powerful that they cannot be stopped or reversed anymore. So this is the idea that there's going to be a point in time when machines are going to become more powerful, more intelligent than humans, and they will take over the world, and they will enslave us or any other thing like that. But what I want to emphasize exactly, AI surpasses human intelligence. That's exactly what we mean by the singularity. And what I want to say is because we're nowhere near approaching this moment of the singularity, no one needs to worry about the fact that artificial intelligence will take over the work of data scientists. Why? Because no matter how advanced algorithms are today, they are still advanced in sort of a very finite set of problems that they can solve, which is called narrow AI. So each algorithm can solve a very particular set of problems, like Siri, for example, can understand voice commands. That's what Siri is good at, but Siri cannot cook, right? Siri cannot do the dishes. Similarly, the algorithms that Cambridge Analytica used to predict our personality were very powerful in terms of predicting our personality, but they could not, for example, predict stuff like how many steps you did today, all right? So what we mean by narrow AI is that we have algorithms that are trained to accomplish a very specific, narrow set of tasks. And for the singularity to occur, we need to have general AI. We need to have robots or algorithms, cyborgs that are capable of performing all of these things that humans can, that can read and hear and speak and walk and do all of these things. But we're very far from approaching this. So there is no need to worry that AI will take over the work of data scientists, because there's no AI at this point that can make better judgments about very complex questions than the human mind. And what I mean by this is algorithms are terrific when it comes to answering questions such as, what's in the data, sort of the descriptive nature of things? What is this data telling us? Algorithms can help us a lot with that, but algorithms cannot answer a more pivotal question, which is why do we observe these things in our data? Okay, so artificial intelligence cannot explain why something is occurring. It can just point out to us that it is, in fact, occurring, not explained why. So we're near the end of my presentation, but I wanted to sort of demonstrate to you some exciting new developments in data science. Has anyone heard of computer dreaming or deep dreaming before? I'll show you, okay, so some of you have already heard of this, but I will show to all of us what exactly this is. And it's based on something, sort of a branch of artificial intelligence and data science that we call computer vision. So let me just demonstrate to you in about 30 seconds what exactly computer vision is. All right, so I hope everyone can see my screen. And I want you to pay close attention to the faces of these people. Does anyone remember that show? Yep, exactly. It's friends. So let me just play it again, just in case you didn't catch that. But did you notice anything sort of out of the ordinary? Yep, that's exactly right. So if you pay close attention to their faces, you will see that it's actually the exact same face on. Yep, so Monica definitely looks different because it's not Monica, but in fact, Nicholas Cage. Yep, so all of them have the face of Nicholas Cage. So isn't that interesting? So how exactly can we accomplish that? It's not Photoshop, right? Photoshop works for sort of static images, but this is an entire video of people that are moving, right? They're moving around, they're saying stuff. And yet it's very believable that they all have the same face and it's the face of Nick Cage transpose on all of their faces. So the technology that makes this possible is called computer vision. Yep, so the apps that people have heard of that are capable of these things, they operate on this principle of computer vision, which is the branch of data science that teaches quote unquote, algorithms how to see things, how to detect, in this case, faces. So basically what the algorithm is doing is first it's mapping all the faces of the people that are in the frame. And then it sort of cuts out the actual face. And it superimposes Nick Cage's face on top of it. But for the algorithm to be able to do that, first of all, it needs to be taught how to recognize what a face is. And then based on the same idea, we have something called computer dreaming or deep dreaming that was developed by Google. And I apologize for the ads, but we'll be able to see the video or sort of an example of computer vision in just a second. And this is what computer vision can do. So it's an art form, almost, of taking an exorbitant amount of data and visualizing this data in interesting ways. And so here as well, we have the advances of data science in the sense of computer vision and artificial neural networks, being able to create something that looks very much like art. And what I want to mention is that if you follow this link that I'm going to drop in the chat, it's called deepdreamgenerator.com. You should be able to create your own sort of artwork based on machine learning. It does require a Google account though. So if you don't have a Gmail account, you can create one and use it. And then you can sort of create your own art based on artificial intelligence without any need for coding or anything like that. So these are just a couple of interesting developments in the field. But I want to conclude by showing you this slide. Does anyone know who this is? Have you seen her before? Or maybe heard of the story? Does anyone know why she's holding a white mask? So does anyone remember her name? Yep, so exactly. So this is one of the leaders in the movement for social justice when it comes to AI. Yep. So there are many documentaries that feature her work. Her name is Joy Boulombini and she's at MIT. And as a PhD student there, she discovered that a face recognition app could not recognize her face. It had no problem recognizing sort of mapping, detecting the face of her white colleagues. But when it came to her face, the algorithm simply could not detect it. She even found out that if she was wearing this white mask, the algorithm could still detect the face but could not detect her own face. And this is when she started really researching deeply how bias occurs in algorithmic systems. So what I want to say about this is that computer science programs are really great at teaching students how to create algorithms. Business programs are really great at taking these algorithms developed by computer scientists and applying them to sort of enhance the bottom line of companies. But what we at the iSchool are great at is going beyond these two domains and also talking about the social implications of algorithms. Okay. And this is something that you don't get to hear about in regular computer science programs. But we do have entire courses that are focused on examining sort of the positive, but also the negative effects of algorithms and how we can improve those and how we can create a better future for everyone, especially marginalized communities. So the last thing that I want to do is reintroduce you to this question. I know that you've seen it before, but I just want to check if this lecture helped you think about data in a different way. And then I'll be happy to answer any questions that you might have. And yes, it is a great documentary. There are other documentaries like that. I can provide you with some more resources if you're interested in sort of this discussion of the social impact of AI. Yep. Okay. So I can do that. Another good one is the great hack. I don't know if people have heard about it. It was also on Netflix at least last month. Okay. Yeah. So that's another good one. The coded gaze is another one. Or coded bias, I believe. But at any rate, I'll create a list for all of you who are interested in this topic. All right. So I will just take a brief look at the answers and then we'll open it up for questions. All right. So we have future valuable statistics, personal interests. Yeah. So what I hope to also spark is sort of this discussion around ethics and justice. And we can talk about it more in the Q&A session. So this is all I have for you for tonight. And now I'll be happy to answer any questions that you might have. I'll stop sharing my screen and I'll pass it back on to Ian. Wow. Thanks a lot, Jess. You know, that was a very interesting look at data science as a whole, but really how it's utilized not only in businesses, but across all different types of applications and how our students can kind of see themselves moving forward in that. It was very interesting. I want to thank you very much for taking time this evening to share all of that information with us. With the 10 minutes or so that we have left, we'd like to open it up for any questions that people have. So please feel free to type your questions in the chat. Jasmine can read them and answer them or if you'd like to raise your hand and say your question over the microphone and even your camera, that would be fine as well. And again, thank you, Jasmine, for sharing everything with us tonight and feel free to ask any questions that you have. I see a lot of thank yous. I will say this, this session has been recorded, so I'm putting my email address into the chat right now. If you're interested in rewatching this recording or if you have questions for Jasmine that you'd like me to pass along, we can answer them that way. I did see a question go by real quick in the chat. I can find that. Do you have majors dedicated to the analyst side of data science? As a matter of fact, new this fall, we have a new major at the iSchool called Applied Data Analytics, which would allow students to explore these cutting edge concepts like machine learning and artificial intelligence. You will take a deeper dive into coding on that. So as a matter of fact, as a matter of fact, we do have a new major that's designed just for that. With Data Analytics as well, is it offered as a minor? So we do also offer Data Analytics as a minor. That would be really reserved for students outside of the iSchool because of the majors we offer. There's a lot of overlap and you wouldn't major and minor in the same thing, but we do offer Data Analytics as a minor at the iSchool. Are there any questions for Jasmine, before the session ends about the the presentation that you saw tonight? If I were to major in DJ and wanted to do a dual major with you. Can you just describe to me what BDJ means? Digital journalism, I would imagine is what you're saying. So if you were to major in digital journalism and wanted to do a dual major, what would be a good major to study that would help you? So we do actually offer dual degree programs with SI Newhouse School of Public Communications where you would pick any major at the iSchool and any major at Newhouse. And that's more of a streamlined process for students. And you probably would major in whatever technical aspect you kind of saw yourself drawn to here at the iSchool. Given this topic, I would imagine that applied Data Analytics would be a good fit for you. But say I'm happy to connect with you individually and discuss how that program works and discuss all of our majors, if you'd like, feel free to send me an email at your convenience and we can set up a time to talk about those programs more in depth. So Jasmine, maybe this is one you can answer. Someone had asked where you could work after getting a degree in data science. So there is no one way to answer this question, but what I could say is that I have tremendous hope in the future in your generation, really, and your interests. And I believe that since so many of you have already seen code advice and other movies and documentaries like The Social Dilemma and all that, and ethics is something that you do consider important. I see nothing but sort of optimism in the field right now and a lot of energy around data ethics, data justice, and so on. There is a lot of work that's being done in this domain right now. And I believe that with all of this, with a collective effort, the field will be tremendously improved when it comes to protecting our privacy and so on. But if this is the kinds of improvements that you were asking about, of course, if you mean on the tech side, I believe that we see improvements every day almost. So algorithms and their power really increase exponentially. And so every year there's a more powerful tool that's out there. But what I want to emphasize is that what the iSchool can give you is not just sort of a state of the art overview of what's available in the field right now, but also equip you with the analytical tools and the critical thinking skills to keep up with these advancements as they keep developing and entering the market. In reference to the question about where students could find themselves working with a degree that focused on data science or data analytics, correct me if I'm wrong, Jezmeena, but honestly, there aren't any fields that aren't utilizing these skills. Currently at the iSchool, we have a 90% placement average. That's our three-year rolling average. We were 88% at an 88% placement rate during a pandemic last year, which is to say that last year 88% of our graduates went on to find employment within six months of graduation. The reason we're able to have such success for our graduates when it comes to job placement is that we're not sending students, we're not sending graduates to one or two fields. We send graduates to every field, right? Every industry utilizes these skills. So as far as career outcomes are concerned, we really have the opportunity to pursue fields that you're passionate about or that you're very interested in. With the few minutes that we do have left, I want to encourage you to ask any last-minute questions you have, but again, you can send those questions to me. I can curate them and send them to Jezmeena and have her answer them, and we can get them back to you. Do you think having a minor in data analytics is useful to those interested in business as well? I would say that it's certainly a very good idea and that's kind of a combination that you'll see a lot of students from Whitman School of Management, the Business School at Syracuse University, will often major in that data analytics here at the eyes or minor in data analytics here at the iSchool because it does complement that curriculum so well, and it offers kind of an added skill set to those students that are looking to enter the classic business roles for organizations. That's a very common combination, as a matter of fact. I want to thank each and every one of you for taking time out of your evening to join us and to listen to this presentation, and I want to invite all of you to check our events page, which I had listed earlier in the chat. It's at our website where you can find all of our upcoming events, including information sessions and other events designed to give you a better idea at the curriculum and the programs that we have here at the iSchool. As I said before, you can reach out to me directly and I'm happy to help you find the resources that you're looking for or get any of the questions that you have answered. I want to thank Professor, Assistant Professor Jasmine Atakeva again. It was very interesting. I'm very excited to have these opportunities to show off our faculty and the incredible work that they're doing. So thank you very much. I'm very grateful for you sharing your evening with us as well. Thank you for this opportunity and thank you everyone for being such a great audience and I really am really pleasantly surprised at how active you were and all the great answers that you were giving me. And so I would just like to put my email in the chat too, in case you have questions, feel free to reach out to me as well. And I hope to see you at the iSchool soon.