Focused Concept Miner (FCM): Interpretable Deep Learning for Text Exploration – MSI

Working Paper

Focused Concept Miner (FCM): Interpretable Deep Learning for Text Exploration

Emaad Ahmed Manzoor

Carnegie Mellon University

Zhaoqi Cheng

Carnegie Mellon University

Dokyun Lee

Boston University

Apr 28, 2021

Introduces the Focused Concept Miner (FCM), an interpretable deep learning text mining algorithm to (1) automatically extract interpretable high-level concepts from text data, (2) focus the extracted concepts to correlate with user-provided business outcomes, and (3) quantify the correlational importance of the extracted concepts.

Topics

Consumer Behavior & Insights marketing analytics

research recap

Opening the Black Box with Interpretable Machine Learning

Machine learning can detect useful patterns in unstructured data and augment or automate managers’ decision making. Unless results are transparent and explainable, however, the use of such algorithms may perpetuate unintended and unseen biases that erode public trust and acceptance. Interpretable machine learning can address these shortcomings.

By using MSI.org you agree to our use of cookies as identifiers and for other features of the site as described in our Privacy Policy.