Tyler Miller Research Notebook

[tmiller-10]

This is the start of my research notebook for my senior comp project

[tmiller-10]

9/11/23:

This was my first meeting with my first reader, Professor Doug Luman. In these posts about my research notebook, he will be referred to as DL. We had numerous conversations regarding my project. After a deep dive into it, I discovered that some implementations are needed for my senior comprehensive project. When Elon Musk acquired Twitter, he rebranded it by calling it X. My main idea was to pull tweets from X(Formally Twitter) from the data source of Twitter using the academic platform that Professor DL. had within the college. When looking into X's version of this academic platform, it told me that it would only be able to output 1,500 tweets a month, which is not enough for what I may need regarding the types of logos. As I continue my research, I will look for a way to verify the data source of Twitter 1.1. This way, the researcher is able to collect the data source, but if I use the Twitter 1.1 source, then I will have to have multiple sources of data to have the company and logos tweets pulled.

[tmiller-10]

9/25/23:

This meeting with Professor DL, We talked about how I need to make a new repo for a particular set of secret keys that are in a safe position. In this meeting we also talked about how it would be feasible to have 1 logo regarding the demo ,but insert historical data from the datasets API. I would need to include some search functionality so that I can pull up the logo within the historical and combat the historical data within the logo changes. Professor DL and I discussed that if there is no other way of getting data that I can use twitter data and then also apply for the 1939 fund within the college. My goal is to find a database or have to pay for the data using the 1939 fund from Allegheny College.

[tmiller-10]

9/29/23:

Today, I attended the Computer of Information Science Alumni social. During this I was able to discuss and talk about a lot of criteria of computer science. I met Rachel Harris who was a class of 2023 and she did the 3-2 program with CMU. She told me that the program is an excellent way to excel in the field of computer science and absolutely loves the school. I learned that program is able to better your knowledge on just about everything related to computer science and how to better yourself for the job search after graduate school.

[tmiller-10]

11/15/23:

I have progressed much on my Senior Comprehensive Project since my last post. With that being said, here is my latest work for my project. When meeting with Professor Luman, we discussed how Professor Ishita Sinha Roy would also be an excellent resource for my comp topic. When meeting with her, she talked about how she found my topic an interesting one for sure. An example we discussed was the Burger King logo and how I could reach out to JKR Global, Jones Knowles Ritchie, an advertising marketing company. She brought up a good point that I need to make sure to be on the lookout for sentiment values that could get confused that are based on the generation words, like words that our generation uses that you would think to be related to a negative connotation, when in reality our generation uses positively. Professor Ishita Sinha Roy also stated that the tool's goal should be a way of tracking the change and correlating it with not just impact but the emotional impact for the good of the brand. With these particular meetings, I have gained a lot of knowledge from my resources, and I can't wait for future meetings with these two great Professors.

[tmiller-10]

2/13/24

As a few months passed, I researched more ways to conduct experiments to get results for sentiment analysis. During winter break, I conducted more research on web scrapping. After further research, it was rather trickier than we anticipated. When getting back from winter break, Professor Luman and I met for an office hours appointment, and we decided for the timeline of the project to do synthetic data for my project and build a database for the synthetic data. When getting this data, I will be using ChatGPT to get responses. I will create three fake case studies that will be surfacing trends within the logo change from that company. Basically, I will create a scenario for a fake scenario for the company, and then I will pull the generated tweets from the positive, neutral, and negative responses. I will be placing these scenarios within my methods chapter of my senior comp. I will also be showcasing a graph of the results. It will be like a poticial graph at the end, I will be looking into the graph more as the weeks go on.

[tmiller-10]

2/20/24:

I recently had a meeting with Professor Luman on the next steps that I had to perform for my methods chapter and I will be performing 3 case studies within my sentiment analysis tool and I am going to be writing those within the next few days as that will be a final version of my chapter for my methods. The link below is the in-class activity on when I looked at a 580 student department project.

https://github.com/CMPSC-580-Allegheny-College-Spring-2024/exemplarproject-TitusSmith33/issues/3

[tmiller-10]

2/24/24:

After meeting with my first reader, Professor Luman, and the rest of my comp group yesterday. We talked about during our experiments chapter of our comp to describe the process and results that we have been able to perform within our testing. We also talked about how this could also allude to the future work of our project and how if someone were to also pick up the project, what would be some implements that they could potentially change, or other situations that could overall change the implementation of the project.

[tmiller-10]

3/13/24:

For our class assignment, we were to make an issue for the README file. I made a Github issue for Titus Smith. Here is the link for that Github issue.

https://github.com/CMPSC-580-Allegheny-College-Spring-2024/exemplarproject-TitusSmith33/issues/4

[tmiller-10]

3/15/24:

Within the current state of my project, I have finished the writing portion of my senior comprehensive project. I have also finished all of my data portions for my data proportion. All I have to do now is polish up my database structure to pull out some portions of tweets that I cannot pull out from the data.