Blog #7: Surprise!

Sai G -

Understanding the Effects of Media Sentiment on Stocks and Businesses

Hello again readers, and yes, it is that time. I know I promised that I’d unveil the surprise last week, and I’m here to deliver. 

But a quick life break before that. I am feeling much better and much more motivated since last week’s illnesses. I’ve managed to make a steady recovery, and I’ll be working on gathering my data for my remaining two data sets for Prime Hydration and Kylie Cosmetics. I’ll have to gather the publication dates and the polarity and subjectivity scores from each article. It will definitely be a time-consuming task, considering how long it took to just get the first dataset done, but I’ll hopefully be able to be much quicker this time around. Plus, since I’m really excited to move on to the next stage of my project, I’m even more enthusiastic to finish this work.

Now, on that note about my next stage of the project, the surprise I’ve been holding back from all of you is …drumroll please… a new coding algorithm. Hopefully, that surprise was as exciting for all of you as it was waiting for it. I’ve decided to develop a new code to get this information on the change in stock prices because of the time-consuming and laborious process that was to gather the sentiment scores from each article. So, this time around, I’ll be opting instead to develop a code that will be suitable for this process. And as you all know from my previous trials with coding, I am not great at coding or anything CS-related, so I’ll be once again learning how to work this out and getting some help from my friends. I just hope to finish this side quest quicker than last time. Now, delving into the purpose and significance of the code, it is necessary for this step of the process because it not only ties my findings to the business’s success but also because it will make the process much more efficient. So, this will be a vital step, but it will mark the close to my data collection, and then I’ll be able to move on to synthesizing the data and results. 

I hope you enjoyed this week’s updates, and stay tuned for more on my awkward tangos with CS. See you next week. 

More Posts

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.

    aryan_r
    Great work Sai! I hope to use your algorithm in the future! What specifics topics do you need to know within CS in order to be able to create your final product? - Aryan
    April 11, 2025 at 4:16 pm - Reply
    sai_g
    Thank you, Aryan! I’m mainly focusing on Python basics and natural language processing—specifically sentiment analysis with tools like TextBlob. I’ve also needed to learn a bit about web scraping and data cleaning to build the usable final product. Appreciate the support!
    April 12, 2025 at 12:58 pm - Reply
    rehan_n
    This is such an interesting project, Sai! I’m curious—how do you plan to measure the impact of sentiment on businesses like Prime Hydration and Kylie Cosmetics, especially since their success might not be directly tied to stock prices? Will you be looking at other metrics like sales data or social media engagement?
    April 13, 2025 at 5:35 am - Reply
    Anonymous
    Hi Sai, this project is looking great? How are you going to quantify your results with sentiment analysis? What units will be used, and what do they mean?
    April 13, 2025 at 5:35 pm - Reply
    jae-hyeok_m
    Sorry about that this comment above ^ was mine, I got logged out somehow in the middle of writing it. Also that first question mark should be an exclamation point!
    April 13, 2025 at 5:37 pm - Reply
    sai_g
    Thank you! That’s a great question. Since Prime Hydration and Kylie Cosmetics are under parent companies, I'll be looking at their stock performance to measure business performance. While it’s not a perfect representation, it does help show how media sentiment can influence broader financial performance. I’d definitely love to explore metrics like sales data or social media engagement in the future, but stock price correlation is the most accessible and consistent metric I can analyze across all three cases right now.
    April 16, 2025 at 9:54 pm - Reply
    sai_g
    You're all good, Jae! For sentiment analysis, I’ll be using two main scores: polarity and subjectivity. Polarity ranges from -1 to 1, where -1 is a very negative sentiment, 0 is neutral sentiment, and 1 is very positive sentiment. Subjectivity ranges from 0 to 1, with 0 being very objective (fact-based) and 1 being very subjective (opinion-based). I’ll be averaging these scores across articles and relating them to changes in stock prices to see if more positive or negative sentiment corresponds with rises or drops in stock performance.
    April 16, 2025 at 9:56 pm - Reply

Leave a Reply

Your email address will not be published. Required fields are marked *