Week 5: Onto Stage Two: Diving into Sentiment Analysis
Krish S -
Well, it took awhile, but I am finally done with the film review collection part of my research project. The most frustrating part of my research is finally behind me. While this stage did take longer than expected, ensuring consistency and quality in my sample was crucial, so I am glad that I took the extra time to refine my approach. I have already set up three data tables (one for each news outlet), and I am ready to conduct sentiment analysis on each of my 120 film reviews.
Looking at the different sentiment analysis tools, the more complex it is, the more accurate results I will get. However, I am not built to code highly complex models, so I will be sticking with simpler models that will still give me accurate results (just not as accurate as those crazy complex models which some articles rave about). I have been using a lot of help from VADER (Valence Aware Dictionary and sEntiment Reasoner), which is a good sentiment analysis tool for the analysis of film reviews. It gives an overall sentiment score that ranges from -1 to +1, which is easy to compare across reviews. I would say that I am around 75% done with coding the model. Based on that, I’ve set a realistic goal of finishing my whole data collection process by next week.

Comments:
All viewpoints are welcome but profane, threatening, disrespectful, or harassing comments will not be tolerated and are subject to moderation up to, and including, full deletion.