Textual analysis7/31/2023 ![]() ![]() #> 5 OnePlus~ Design, Features, and~ "There are two speaker grilles flanking the m~ #> 4 OnePlus~ Design, Features, and~ "The 5.5-inch, 1080p IPS display is on par wi~ #> 3 OnePlus~ Design, Features, and~ "Our white test unit features a so-called sil~ #> 2 OnePlus~ Design, Features, and~ "The OnePlus One doesn't feel like a sub-$400~ #> 1 OnePlus~ Introduction "The days of the $600 smartphone aren't over ~ The structure of the data can be seen below: head(reviews) #> Pass a specification to the `col_types` argument to quiet this message #> Use `spec()` to retrieve the guessed column specification # You can download the data using this link # In this case we will do it from my GitHub repository ![]() # We will need a URL from where to download the data You can learn them just so you can allow yourself to be too lazy to read a book or a review front to back. See, these text analysis skills can be used for selfish reasons as well, it doesn’t always have to be something “useful” or “productive”. I want to check what is the general opinion about them and see if my decision was right or was it just bias. I could have chosen any other brand of phone, or any other product for that matter, however I own a OnePlus. As a dummy dataset I have chosen a series of reviews for the OnePlus phone models. So, with the disclaimers out of the way we will discuss:Īs a dataset, I though that a series of phone reviews would be a good starting point.Īs an analyst, it might be required of you to spend some time analysing reviews for different products your company makes and get insight from said reviews. That being said, with some tweaking, they can reliably be used as a starting point in the endeavours to automate this process, with more complex methods being added at a latter time. I would also like to point out that I will show some basic sentiment analysis methods, however they do not cover all the possibilities and are just the tip of the iceberg. Here, I will just show you three methods that can cover a great deal of analytical needs in a company. It is not intended as a comprehensive course on Natural Language Processing (NLP), as that is a complex topic that cannot be dealt with in just one article. This article is intended just as an introductory example into what text analysis can do and how it can be used by data analysts, although I encourage you to study further if you think these methods can be useful. I would like to shortly discuss the structure of the article and make some disclaimers about it, so we are on the same page. I highly recommend this book as their approach is to transform the text into a tidy format that allows you to easily analyse and visualize the results using graphs. So, in order to see how to analyse text using R I have started reading Text Mining with R by Julia Silge and David Robinson. Now, what about the second case, when you have a one time report? Wouldn’t it take just as much time to set up the report as it would to analyse it manually? Well, no and you will see in the article below how easy it can be to analyse data using R. The other solution is to analyse the data manually every time. You need to spend the time to set up the analysis, graphs and report, but this needs to be done just once and you can use it every time you want to refresh the report. In the first case is kind of self-explanatory. In both cases text analysis can be very beneficial. You can be required to analyse text either as a recurring report or as a one time analysis. You can avoid spending the huge amount of time on this situation if you learn how to analyse text using a programming language and here is why. The way I used to do it was to read the comments, categorize each one then do some basic analysis based on those categories. As you might know, if you’ve done this sort of analysis, this can be a really boring and tedious work and can take a large amount of time. I always hated the annoying task of having to analyse text, be it in the form of comments on the internet, transcripts from focus groups or something else, does not matter. So I thought to myself, what would be a good skill that would help me as a data analyst or would have helped me in the past? It had to be something that took a lot of time and that could be automated. Recently I found myself with some free time on my hands so I decided to learn a new skill, or at least start learning. ![]()
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |