Project Title: Coding ML Algorithms to Highlight Major Real World Issues
BASIS Advisor: Matt Ramsby
Internship Location: Plus3 IT Systems
Onsite Mentor: Kevin Hulsing
Machine learning is a branch of AI that uses patterns and references to categorize data by itself. This project will dive into a problem that plagues machine learning programs: bias. Specifically, algorithmic bias, which is a case where the bias comes from the machine itself, rather than the reference data set or the programing. Algorithmic bias is extremely hard to get rid of in a machine learning program and can cause havoc in peoples' lives if it exists within certain programs, such as job searching or targeted ads. This project will result in a literary review on the problems of algorithmic bias along with cases in the past involving algorithmic bias in the real world and the problems it caused. To help in this project I will be working with Kevin Hulsing from Plus3 IT Systems. Our goal is to induce algorithmic bias in a machine learning algorithm in order to further the knowledge of algorithmic bias in scientific community.
My Posts
Week 10
This week has been a generally boring week in terms of my senior project. Since my project was wrapped up last week, this week has been full of me working on my presentation. However, there were some other interesting things happening that are not related to my senior project. I got my second shot and […]
Week 9
This week, I sat down with my advisor in a Google Meet and we worked out how to deploy my Machine Learning model to the cloud. To do this, we use Microsoft Azure, which is a cloud computing tool that can do much more than deploying models, but that’s all we need it for. The […]
Week 8
In this past week, We have been exploring how to modify the machine learning algorithm correctly so we source can be taken into account when an article is being run through the program. We have not been successful and have decided to move on to the next topic. the next topic is expanding the model […]
Week 7
This week, we attempted to modify the machine learning model so that it would include the source of the article when categorizing the article as fake or real. We also changed the Real/Fake tag to Biased/Not Biased as to make it more neutral and objective. we ran into some problems with the modification though. The […]
Week 5 and 6
These past few weeks have mostly consisting of running code every day to scrape the headlines and articles off of CNN and FOX news websites. I have collected around 1500 articles over the past two weeks and I am starting to move onto the next bit of my internship. The next part include scraping a […]
Week 4
This week we started modifying the base code we were working with. I built a Web scraper for Foxnews.com that reads todays headlines and pulls the content out of all the top articles. The content is then run through the Machine learning model to see if it is fake or real. I am working on […]
Week 3: Starting the Code
This week, I made a working model for a machine learning algorithm that was able to tell the difference between real and fake news. It is incredibly bare bones right now and needs some modification for it to be useful, but it is a strong start to the bulk of my project. In this past […]
Week 2: General Pain
This week was eventful. I terms of my project, I researched the terms of what I will be using in the coding section of my project. The two big ones were: TfidfVectorizer: this used term frequency to build a vector using the term as the direction and the frequency of that term as the magnitude. […]
Starting my Project
Hello everyone, This is the first week of my senior project. I am working on creating a machine learning algorithm to detect fake news. In the first two weeks, I will be working on learning the terms and vocabulary of machine learning programs, in order to code and discuss the project freely. I will also […]