dzianis-pirshtuk Goto Github PK

attention-sentiment

My bachelor's degree thesis (with code and experiments) on sentiment classification of Russian texts using Bi-RNN with attention mechanism.

classifying-twitter-user-as-resident-or-tourist

Researches confirms that social media provides good insights on what people think, feel, concern, etc. It is expected that those insight mined from Twitter data has potential to support a better decision-making, especially in public sectors. Public sector wants to know local’s insight level; therefore they need to make sure they use the conversation from residents. However, the ground truth shows that tweets are mixed from the residents and tourist. This study investigates the best automatic fashion model to classify tweets posted by resident and tourist, in NTB. Indonesia. To do so, several consecutive phases were conducted. Those are pre-processing, data training, classification system, data testing, accuracy comparison, and result visualization. First of all, a Twitter dataset, which has 700,000 tweets posted by approximately 26,000 users in Nusa Tenggara Barat, Indonesia was prepared. The dataset divided into two sets, tweets from 4,000 users for data training and 22,000 users for data testing. Then, three popular classification algorithms were applied to the datasets. There are Multinomial Naïve Bayes, Support Vector Machines and Decision Tree. After that, 7 features are created. There are Bag of Words, Normalizer location, Total Tweet, Total Day, Tweet per Day, Total Location and Location per Day. Experiment shows that Multinomial Naïve Bayes with Bag of Words feature has 86% accuracy, while the rest of features give less than 65% accuracy. This is different with Support Vector Machines and Decision Tree results. These two algorithms produce better accuracy results excluding Bag of Words feature. It implies that Support Vector Machine and Decision Tree are more powerful when processing numerical value. However, among all classification system, Multinomial Naïve Bayes still being the most accurate algorithm for the model.

conda-tensorflow

conda_auto_activate

Automatically activate a conda environment when entering folders/project.

conference

A WebRTC signaling server with support of MQTT and WebSocket as transport protocols, token based authentication (JSON Web Token) and external policy based authorization.

data-analysis-course-bsu-mmf-2016

data-analysis-course-bsu-mmf-2017

datathon2017-masterclass

Example of solving a problem in predictive analytics for marketing (banking case)

dato-core

The open source core of the GraphLab ML library

deep-face-generation-and-editing-a-survey

Deep Face Generation and Editing: A Survey

dest

:panda_face: One Millisecond Deformable Shape Tracking Library (DEST)

display-advertising-challenge

Criteo/Kaggle Competition of CTR prediction

dynamic-customer-targeting-in-r

dzianis-pirshtuk.github.io

face-landmark

caffe 68 points face-landmark

fbtextclassifier

EECS 498 (Intro. to Information Retrieval) Final Project: Text classification applied to social media

gender-detector

Library for guessing a person's gender by their first name.

geonamescache

geonamescache - a Python library for quick access to a subset of GeoNames data.

gitlabhq

GitLab is version control for your server

harvester

The Social Harvest server that exposes an API and harvests data from the web to be analyzed.

kaggle-2014-criteo

kaggle-criteo

Kaggle Criteo https://www.kaggle.com/c/criteo-display-ad-challenge

kaggle-galaxies

Winning solution for the Galaxy Challenge on Kaggle (http://www.kaggle.com/c/galaxy-zoo-the-galaxy-challenge)

kaggle_criteo

Software for the kaggle criteo challenge

kphp-kdb

VK-KittenPHP/DB/Engine suite

loan-default-prediction

Loan Default Prediction at Kaggle

mlg.instagram.dataset

Code for building a co-occurence data set from instagram

netflix-prize

The code I used to get in the top #150 in the Netflix Prize

onnx

Open standard for machine learning interoperability

polyglot

Multilingual text (NLP) processing toolkit