All right, so what we're going to be taking a look at in this course is social media monitoring platform. We're going to be using Crimson Hexagon as a platform that we conduct our work in. So we're going to focus first on doing social media monitoring using the platform. We'll go through the metrics that are reported, the types of monitors that we're able to set up. And then we're also going to get into, in a subscript module, doing the analysis of the social media data. So the data that we're able to export from Crimson Hexagon, how can we analyze that data using statistical methods? So what you see on the screen right now, this is the screen that lists all of the monitors that are actively running within Crimson Hexagon. And there are three types of monitors that we are going to take a look at today. We're going to start with the buzz monitors. Buzz monitors use the automated tools that are built into Crimson Hexagon's fore side platform. So it's going to provide us with automated sentiment analysis. We're going to be able to do a little bit of content analysis, gain some information about the contributors of the social media content. We'll also take a look at opinion monitors. Opinion monitors allow you to train the algorithm, so this has been seen to be more reliable across different contexts because rather than relying on the automated algorithms, you're going to tailor it with human judgment. And then the last monitor that we're going to take a look at are social account monitors. This lets you take a look at what is being posted from particular social accounts. And so the example that we're going to use for all three of these is doing analysis of the Starbucks brand. So I'm just going to scroll down, and you'll see the three monitors that we've created for Starbucks already. The first monitor that you see labelled Twitter, that's the social account monitor of tweets coming from the Starbucks account. The second monitor you see is the BUZZ monitor, and then lastly we have the opinion monitor that we're going to train. So let's start by taking a look at the BUZZ monitor. I'm going to click on Edit so that you can see how you go about constructing these monitors. And one of the nice things about Crimson Hexagon and other social media monitoring platforms is that they've made it really easy and accessible, as far as constructing the monitors, setting the date range, specifying any constraints that you might have. And so I'll click on Edit so you can see the construction. And these have already been created to save us some time for the data to be pulled. But you'll see up top, we named the monitor. The first set of options for us is to select the content sources. Now you'll see that I've already selected Twitter, Facebook, blogs, and discussion forums to get a range of content. One thing to note, that Crimson Hexagon and other social media monitoring platforms. It's going to be pulling in the public social media data, and so unless someone has specified their account to be private, we're going to pull in all of their Twitter posts related to a particular query that we run. Now that's going to probably stand in stark contrast to Facebook posts, the majority of users there have their accounts specified with privacy settings. And so we're not going to be able to pull as many posts from Facebook. The way that this is going to be reflected in the data that we pull is across a number of contacts, it's largely going to be dominated by conversations pulled from Twitter. And so that's one thing that we do want to keep in mind. You can see there are more advanced options that we can specify, particular pages that we want to pull content in from. We might choose hashtags from Instagram. We might say that we want to pull that information from news or product reviews websites, really depends on the context that you're looking at. But for demonstration purposes, we're going to focus on these first four content sources. So once we specify the content sources, we're going to specify what the phrase is that we're searching for. Now Crimson Hexagon offers two different ways to go about doing this. One is to use their guided search. And so we can specify a query based on all of the words that we want it to include. So you can see this query is very simple. We're just using the term, Starbucks. So if I were to enter multiple terms, so if I were to enter, for example, Starbucks, And Frappuccino, And I may be spelling this incorrectly, but if this were the query that we were to run, Starbucks Frappuccino. What this would do is require that posts contain the words Starbucks and Frappuccino. All right, now that's going to be a a little bit different than if I were to include quotation marks around this. And now it's looking for the exact phrase, Starbucks Frappuccino, with those words appearing together. So, very much the way that you would construct a query for a search engine. Now a little bit different would be if I were to put these words in the next line that says at least one of these words. What this is going to do is pull all mentions, all social media posts mentioning Starbucks or the word frappucino. And we also have an exclusion list, right? So it makes it very easy as far as constructing your query using this guided format. I actually prefer using the free form tool, and in doing this, we would use Boolean logic. And you see there's a help guide that's provided, I'll open this up in a new tab, so that we can take a look at it. But what the Boolean tool is doing would be using the phrases and, or using nested statements. And we'll just switch over to that. So you can see the help section that's provided, the operators and, or and with a minus sign for negation. You also see that there's proximity of words. So I want words to show up in close proximity to each other. They don't necessarily have to occur one right after the other, but I want these words to be used in the same phrase. So I might specify that near statement, how close those words can be to each other. Might also specify where we want the comments, the social media posts to be coming from or names of the particular authors. As you can see, you can also specify the influence range. So lots of ability to tailor these queries and combine these search tools together. All right, so whichever approach you are most comfortable with, the freeform, I think, gives you a bit more flexibility. In this case, our query is simple. We're just searching for all social media posts from the content sources we specified that mention the term, Starbucks. Now if we click on validate the query, that will make sure that our syntax has been entered correctly. As we scroll down, we're not going to be searching for logos, but in terms of refining our search results, we can specify the date range. So I've set this monitor up to go back as far as January 1st, 2012. I've restricted it just to English language posts with the location coming from the United States. And we have additional options here as being able to put further restrictions on the results. And we can add white lists and black lists or specific terms that we want to exclude from the sentiment analysis. And then the last feature that's nice, if we go back all the way to 2012, and find out afterwards that we've entered an incorrect search term, that we need to refine it further. It's going to take a while for that data to be pulled. So one thing that you can do is click on Run Preview, and it's going to pull up a sample of the data that will be returned through the query itself. And so this is just looking at the last couple of days. Again, the majority of posts coming from Twitter, very little coming from Facebook in this case. And so that's giving us a reflection of the volume that we see over the last couple of weeks. And we can also see a sample of the posts. And so we see, looks like we've got some product listings, some discussion of the Starbucks at 1st and Pike's at the Seattle location, these coming in from blogs. If we scroll down a little bit further, we get to some discussion that was occurring on other venues, and then we get to Twitter activity around it. All right, so everything looks good there. We would hit Save and ultimately run the monitor. All right, so that's the construction of the monitor. All right, so I'm going to go back to our dashboard and jump into this monitor, so that we can take a look at the results from one of the BUZZ monitors from within Crimson Hexagon. And what I want to do is walk you through the different tools that are built into the platform. There's a lot of insight that we're going to be able to get directly from the social media monitoring platform. And then in the next module, we're going to be using the data that comes from this platform to be able to do some more advanced analysis.
