Comparative Analysis of Topic Modeling Methods

Description

Conducting a comparative analysis of topic modeling methods applied to the task of identifying professional potential in workplace chats.

Motivation

  • Conducting a comparative analysis of topic modeling methods for identifying the professional potential of employees based on workplace chats, with the goal of determining the most effective approaches for further competence discovery.

Goal

  • Perform a comparative analysis of existing solutions for topic modeling of short texts, aimed at enabling a justified selection of technologies for extracting knowledge elements from texts of professional discussions.

Approach

  • The set of workplace chats to be analyzed is not predefined.
  • The solution will be implemented as a set of procedures for semantic analysis of situational discussion texts based on topic modeling methods applied to collections of ad‑hoc discussion texts.
  • The results should enable a justified choice of the most promising topic modeling methods suitable for processing ad‑hoc discussions.

Activity Distribution

  • Programming: 40%
  • Analysis: 25%
  • Theory: 25%
  • Evaluation: 10%

Preferred Skills

  • Interest in NLP, topic modeling, semantic analysis of blogs
Apply