Detecting deception, disinformation, and crowd manipulation on social media through machine learning, natural language processing and artificial intelligence - QC-204

Preferred Disciplines: Computer Science, Machine Learning, Natural Language Processing, Artificial Intelligence. (Masters, PhD or Post-Doc)
Project length: 8-12 months (2 units)
Approx. start date: As long as possible
Location: Montreal, QC
No. of Positions: 1
Preferences: None
Company: Nexalogy

About Company:

Founded in 2010, Nexalogy is a Montreal based company whose goal is to take the guess work out of social data analytics. Their deep analytics platform, called NexaIntelligence, captures structures and visualizes vast amounts of unstructured social data which, as a discovery tool, allows organizations to make better decisions by pinpointing specific trends, risks and opportunities. Nexalogy has a patented method for performing analysis on unstructured information and has a superior data cleaning capability that allows for conversation sculpting.

Summary of Project:

The company has cutting edge ML, NLP and AI stacks that treat social media data in English, French, Russian and Korean. The research to be performed is to automate detection of deception, misinformation, fake news and crowd manipulation in these langauges taking place on social media platforms. The objective is to train models that can be reliabily deployed into chat bot interfaces for augmentation of a social media discovery and analysis platform. The project is to develop and train models, the objective is a chat bot augmentation of the enterprise social media discovery platform called NexaIntelligence.

Research Objectives/Sub-Objectives:

  • ML, NLP, AI for unstructured text disinformation, deception and manipulation detection in French
  • ML, NLP, AI for unstructured text disinformation, deception and manipulation detection in Korean
  • ML, NLP, AI for unstructured text disinformation, deception and manipulation detection in Russian

Methodology:

  • Best in class supervised machine-learning models
  • Social media retrieval, disambiguation, analysis, and automation of entity extraction and extractive summarization
  • Open-interface chat bot implementation on an existing stack

Expertise and Skills Needed:

  • ML, NLP and AI competency including supervised and unsupervised ML
  • Familiarity with integration and deployment of chat-bot functionality to mobile including smartphones and tablets. 

For more info or to apply to this applied research position, please

  1. Check your eligibility and find more information about open projects.
  2. Interested students need to get the approval from their supervisor and send their CV along with a link to their supervisor’s university webpage by applying through the webform
Program: