fptjy Goto Github PK

arkfilter

cs231

Complete Assignments for CS231n: Convolutional Neural Networks for Visual Recognition

cuckoofilter

cuckoopy

Python implementation of Cuckoo Filter data structure

database_networkflow

datamining

data mining class

distance-dependent-chinese-restaurant-process-

Setup to run ddcrp clustering using embedding based distance on a text corpus.

github-slideshow

A robot powered training repository :robot:

jumpfilter

nlp-in-practice

Starter code to solve real world text data problems. Includes: Gensim Word2Vec, phrase embeddings, Text Classification with Logistic Regression, word count with pyspark, simple text preprocessing, pre-trained embeddings and more.

python-bloomfilter

Scalable Bloom Filter implemented in Python

querylikelihood

A toy search engine using Query Likelihood model

shifiing-filter-framework

smoothing-techniques-in-nlp

Some smoothing techniques used in NLP and ML, including Laplacian and Lidstone Smoothing, Absolute Discounting, Backoff, Kneser-Ney and Interpolation.

test

twitter-sentiment-classification-incorporating-users-and-events-

Neural network methods have achieved noticeable progress in Natural Language Processing (NLP), among which sentiment classification of Twitter messages is attracting increasing research in recent years. However, existing neural network methods for Twitter sentiment classification typically only capture the semantics of texts, but ignore the importance of users who express the sentiment and the underlying events on which the messages are commented. In this paper, we propose a comprehensive neural network model by incorporating user information and underlying event information into document-level Twitter sentiment classification. Firstly, users and events are encoded in continuous vector spaces through which important global clues such as user sentiment tendency and event background are captured. Then, the vector representations of users and events are further integrated with continuous text representation in a unified neural framework for exploring Twitter sentiment classification. Multiple experiments have been implemented to verify the performance of sentiment classification on a large-scale Twitter dataset. The empirical results indicate that the incorporation of user and event does improve the classification accuracy and our proposed method yields satisfactory performance.

uis-dat630-fall2017

UiS DAT630 Web Search and Data Mining course, fall 2017

vertical-cuckoo-filter

vertical cuckoo filter

word-prediction-ngram

Next Word Prediction using n-gram Probabilistic Model with various Smoothing Techniques