PROJECTS

DIAGNOSING SKIN CANCER WITH DEEP LEARNING

AI in Healthcare

Skin cancer is the most prevalent type of cancer worldwide and kills thousands of people every year.
In the United States alone, more than 9,500 people are diagnosed with skin cancer every day and more than two people die from the disease every hour. However, it is also often difficult to detect by the naked eye. In fact, diagnosis by medical experts currently only have a 77.0344% accuracy.
In this project, I used current images of known skin cancers and attempted to write an algorithm that detects skin cancer more accurately than unaided visual inspection by physicians.

SENTIMENT ANALYSIS ON INDONESIA'S 2019 PRESIDENTIAL CANDIDATES USING TWITTER API

AI in Politics

The 2019 Indonesian presidential election was a controversial one. It was a close election between the incumbent, Joko Widodo (Jokowi) and Prabowo Subianto.
The result of this election was so close that both candidates declared victory and even though Jokowi was eventually declared the official victor, there are still protests going on urging a recount of the results months after the election results were announced.
In this project, I used tweets pulled from the Twitter API to perform sentiment analysis on the two candidates to see if it could predict the result of the election.

PREDICTING THE AGE AND RELATIONSHIP STATUS OF OKCUPID USERS

AI in Online Dating

In the 21st century, more and more people use online dating platforms to meet future partners. However, people are not always truthful about who they are. In fact, it is quite easy to create a false identity online. 

In order to curb this issue, I analyzed a dataset of 60,000 OkCupid users in San Francisco and used different regression techniques to predict the age of users, and classification techniques to predict the relationship status of users. If this information can be predicted with great accuracy, then it can be used to detect when a user is lying about their age or relationship status.