Robert Ridley's Projects
Alluxio, formerly Tachyon, Unify Data at Memory Speed
This poject is a segmentation tool for Chinese text. It uses a combination of Forward Maximum Matching (FMM) and Reverse Maximum Matching (RMM) followed by a lowest-mean-cost disambiguation algorithm.
Concentration is an ios app. Flip the cards and try to find matching pairs.
This repository contains the code used to produce the results from the paper Automated Cross-prompt Scoring of Essay Traits published in AAAI 2021.
cryptocurrency-tracker is a React Native app. It displays a list of cryptocurrencies along with their current prices and performance over the past hour, day and week.
This repository contains code for three different dimensionality-reduction techniques, PCA (Principal Component Analysis), SVD (Single Value Decomposition) and Isomap. There is also a 1-Nearest Neighbors algorithm which is used to classify the reduced-dimensionality test data.
This repository contains code for LDA (Latent Dirichlet Allocation) for document topic modelling. There exist two different LDA implementations in the repository, a gensim implementation and an LDA written from scratch .
fling-server is a node.js postgraphql backend for a flingapp front-end.
flingapp is a react.js front-end for managing an organisation's book of freelancers.
Repository for Frequency Word List Generator and processed files
A knowledge-base question-answering system for《三国演义》(The Romance of the Three Kingdoms)
This is the code and data used to produce the results from the EMNLP 2023 paper Addressing Linguistic Bias through a Contrastive Analysis of Academic Writing in the NLP Domain.
This NER tagger uses a hidden-markov language model plus the viterbi algorithm to tag Chinese sentences according to whether they contain named entities, such as the names of people, times, places and businesses. The model achieves 96% accuracy and an 87% f1-score on the dev.txt data.
Distributed System, Fall 2017, CS@NJU
This project is a dictionary lookup that will return Chinese translations and part-of-speech tags for English queries. The program also incorporates a rule-based stemmer so that results for words with grammatical suffixes are also be returned. For example, a search query of 'studying' will additionally return results for 'study'.
This is my first foray into the world of AR: a simple game where you tap as many planes as you can in 20 seconds.