Why Didn't You Listen to Me? Comparing User Control of Human-in-the-Loop Topic Models

05/23/2019
by   Varun Kumar, et al.
0

To address the lack of comparative evaluation of Human-in-the-Loop Topic Modeling (HLTM) systems, we implement and evaluate three contrasting HLTM approaches using simulation experiments. These approaches are based on previously proposed frameworks, including constraints and informed prior-based methods. User control is desired, so we propose a control metric to measure whether refinement operations are applied as users expect. Informed prior-based methods provide better control than constraints, but constraints yield higher quality topics.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/04/2023

A User-Centered, Interactive, Human-in-the-Loop Topic Modelling System

Human-in-the-loop topic modelling incorporates users' knowledge into the...
research
09/12/2023

Evaluating Dynamic Topic Models

There is a lack of quantitative measures to evaluate the progression of ...
research
01/06/2021

User Ex Machina : Simulation as a Design Probe in Human-in-the-Loop Text Analytics

Topic models are widely used analysis techniques for clustering document...
research
05/06/2022

Physics-informed neural networks for PDE-constrained optimization and control

A fundamental problem of science is designing optimal control policies t...
research
06/09/2022

Topic-Aware Evaluation and Transformer Methods for Topic-Controllable Summarization

Topic-controllable summarization is an emerging research area with a wid...
research
08/01/2019

Semantic Concept Spaces: Guided Topic Model Refinement using Word-Embedding Projections

We present a framework that allows users to incorporate the semantics of...
research
06/28/2022

A Contribution to the Defense of Liquid Democracy

Liquid democracy is a hybrid direct-representative decision making proce...

Please sign up or login with your details

Forgot password? Click here to reset