Veritas

Veritas is a blockchain repository of trusted news sources vetted through our machine learning algorithms and NLP techniques. We used Taboola's native API to filter trend categories, such as law, politics, etc. and scraped news articles corresponding to distinct topics within each category of interest. We then applied sentiment analysis to articles grouped under the same topic in order to determine a confidence score based on the proportion of related articles that agreed with the sentiment of each headline.

Genome Matcher

Implemented a trie-based genome library with exact and single nucleotide polymorphic search functionality for DNA sequences up to 3 million bases long. Able to quickly identify all genomes in the library within a matching percentage threshold of the DNA of a specified genome. Supports manually input genome/DNA sequences, as well as file inputs.

GASP

Trained a multivariate gradient descent algorithm to determine the probability of getting into graduate school, based on the given parameters: GRE and TOEFL scores, University Rating, Essays, Undergraduate GPA, Amount of research experience, etc. Employed feature normalization techniques to gain an overall training accuracy over 90%. Implemented using only numpy's CSV parsing and matrix multiplication functionality.

sketches and scribbles