Grant: Engage Grants Program
Université du Québec à Montréal
$25,000.00
Bridging Languages in Social Networks and Semi-Supervised Learning for a Compact Representation
FY 2016-17
Summary
Social media is international: users from different cultures and language backgrounds are generating andsharing contents. In the microblogging site Twitter, information spreads across languages and countries.Pre-processing and understanding such information is a real challenge for established companies. Moreover,large-scale text classification and clustering are real challenges in machine learning since labeled data is notoften available or its volume is too small.Advanced Symbolics (ASI) uses Artificial Intelligence combined with online sampling to measure publicopinion on almost any topic. Their patented sampling algorithm, Conditional Independence Coupling (CIC),continuously crawls online sources (Twitter, Facebook, Blogs, etc.) creating a representative sample bygeography (e.g. by country like Canada, United States, by state or province like Ontario or California, or by citylike Toronto or San Francisco).ASI's current Natural Language Processing pipeline is optimized for English and French languages. Thecompany wishes to expand to other languages, such as Japanese, Chinese, and Korean. Acquiring the expertiseto handle this multilingual challenge at this stage at ASI is cost prohibitive; on the other hand the customersasking for this addition would represent a huge win for the company.The current research project aims at creating tools for linguistic pre-processing, analysis, modelling andclassification of Twitter posts, into general categories. In this project, our main concern is the linguisticpre-processing of Japanese, Chinese and Korean short messages (140 characters limit) of Twitter that areunstructured and highly noisy. Moreover, we will develop a classification tool for Twitter posts using the threeAsian languages. As we do not have enough annotated data for the multilingual set of documents,semi-supervised classification techniques will be explored and investigated. The construction of such powerfultool and technology will help design a complete data mining and NLP pipeline at ASI.
Program
Engage Grants Program
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
508048-2016
Name nom
Sadat, Fatiha
Department département
Informatique
Organization id
57
Institution établissement
Université du Québec à Montréal
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2016
Competition year année de concours
2016
Award amount
25000
Program id
EGP
Program name en
Engage Grants Program
Program name fr
Programme de subventions d'engagement partenariat
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1552
Committee name en
Quebec Internal Decision Committee
Committee name fr
Comité de décision interne pour le Québec
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2705
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Software and development
Research subject fr
Logiciels et développement
Partie
1
Nb partie
1
Application title
Bridging Languages in Social Networks and Semi-Supervised Learning for a Compact Representation
Keyword
NSERC
Application summary
Social media is international: users from different cultures and language backgrounds are generating andsharing contents. In the microblogging site Twitter, information spreads across languages and countries.Pre-processing and understanding such information is a real challenge for established companies. Moreover,large-scale text classification and clustering are real challenges in machine learning since labeled data is notoften available or its volume is too small.Advanced Symbolics (ASI) uses Artificial Intelligence combined with online sampling to measure publicopinion on almost any topic. Their patented sampling algorithm, Conditional Independence Coupling (CIC),continuously crawls online sources (Twitter, Facebook, Blogs, etc.) creating a representative sample bygeography (e.g. by country like Canada, United States, by state or province like Ontario or California, or by citylike Toronto or San Francisco).ASI's current Natural Language Processing pipeline is optimized for English and French languages. Thecompany wishes to expand to other languages, such as Japanese, Chinese, and Korean. Acquiring the expertiseto handle this multilingual challenge at this stage at ASI is cost prohibitive; on the other hand the customersasking for this addition would represent a huge win for the company.The current research project aims at creating tools for linguistic pre-processing, analysis, modelling andclassification of Twitter posts, into general categories. In this project, our main concern is the linguisticpre-processing of Japanese, Chinese and Korean short messages (140 characters limit) of Twitter that areunstructured and highly noisy. Moreover, we will develop a classification tool for Twitter posts using the threeAsian languages. As we do not have enough annotated data for the multilingual set of documents,semi-supervised classification techniques will be explored and investigated. The construction of such powerfultool and technology will help design a complete data mining and NLP pipeline at ASI.