Yuki(Yuxin) Chen's repositories
Design-and-implementation-of-a-task-based-Q-A-system-based-on-TCM-knowledge-graph
Cleanse the data of "Compendium of Materia Medica" as the full-text data foundation, with herbs as the center, to construct a Chinese medicine knowledge graph including 6 entity types and various relationship types. Use BiLSTM, CRF, and naive Bayes classification to build a question-answering system based on the Chinese medicine knowledge graph.
CNER-Clinical-Named-Entity-Recognition-
Named entity identification for electronic medical records using HMM
Real-Site-House-Price-Forecast
Crawl the Shanghai area house price information from a house selling website (https://sh.lianjia.com/ershoufang/pg2/) and use multiple regressions to predict house prices
A-hybrid-label-based-Bayesian-personalized-ranking-literature-recommendation-algorithm
This project focuses on the implicit feedback-based recommendation ranking problem, aiming to divide the full dataset into multiple local datasets by item label weights and user label weights, and to use a hybrid Bayesian personalized ranking recommendation algorithm in the local datasets
A-time-series-based-study-of-seasonal-trends-in-flight-passenger-numbers
Based on the monthly flight passenger data of an international flight from January 1949 to December 1960, this study analyzed and plotted various trend graphs of the data, analyzed the seasonal fluctuations and patterns, and studied the establishment and application of seasonal autoregressive moving average models of different orders.
Crawler-projects
A summary of some simple crawler projects, including static and dynamic crawlers
Creditcard_frauddetection_AutoEncoder_proj586
A deep AutoEncoder model is used for credit card fraud detection, which includes a multi-layer network of encoders and decoders and implements the method of reconstructing data to find the error threshold and achieve classification of fraud cases
Data-enhancement-and-paraphrasing-
Research on multi-level Chinese paraphrasing, using seq2seq to generate multiple sentences with the same semantic meaning
DATA551_proj
An interactive dashboard that could investigate homicide-related data from 2010 to 2014 in the US, provide insights into relevant questions, such as examining how homicide-related activity is distributed geographically across the US.
Deep-neural-network-based-knowledge-QA-system-for-the-treatment-of-pediatric-symptoms-in-TCM
Constructing a corpus of ancient Chinese pediatric medicine literature, using algorithms such as BERT, lattice LSTM, and Siamese for tasks such as named entity recognition, intent recognition, entity similarity calculation, and entity linking, to develop a TCM-QA.
Customer-lifetime-value-analysis-based-on-online-retail-data
Using real online retail data from a store, customer value was predicted and segmented using two probability models: BG/NBD and Gamma-Gamma. An example was used to illustrate the implementation process of the customer lifetime value analysis model and segmentation model.
digitalMediathe-Mahogany
the Interactive application designed to promote Mahogany by using Adobe director
github-slideshow
A robot powered training repository :robot:
HomicideDashboard
About An interactive dashboard that could investigate homicide-related data from 2010 to 2014 in the US, provide insights into relevant questions, such as examining how homicide-related activity is distributed geographically across the US.
Income-range-analysis-based-on-personal-information
Using personal information and income range data from the 1994 US Census database, applied naive Bayes classifier and linear discriminant analysis classifier to classify personal income ranges. Through examples, explained the implementation process of the income classification model and the segmentation model.
nlp-notebooks
A collection of notebooks for Natural Language Processing from NLP Town
Performance-evaluation-of-universities-based-on-hierarchical-analysis-and-data-envelopes
Using Analytic Hierarchy Process (AHP) and Data Envelopment Analysis (DEA), a comprehensive analysis of material and spiritual indicators of some universities in Ontario, Canada is conducted, and corresponding performance evaluations of the universities are made.
proj_NLPbrl_DATA534
The NLPbrl wrapper API is a package for wrapping The Rosette Text Analytics API's functions. It can use natural language processing, and machine learning to analyze unstructured and semi-structured text in multilingual. It can provide a large number of powerful text analyses which has largely helped people to learn about natural language processing
RobotSystem
Simple chatbot development in C++
Vulpes-Distribution-in-BC
Using Spatial Statistics methods to analyze vulpes' distribution in BC province.