Christine Murad - LitSense: Supporting Academic Literature Review Workflows

Project Summary

Performed research in the Dynamic Graphics Project lab at the University of Toronto, to develop and evaluate a web application that assists researchers in the literature review workflow
Conducted formative interviews to gather information about academics current methods of performing bibliography searches and literature reviews
Developed LitSense, an end-to-end tool to support literature review workflows from paper discovery to synthesis. Unlike existing reference management tools that maintain "collections" of papers, LitSense is designed to focus on a particular research question, a distinction
relevant to cases where a paper contributes different perspectives to various literature reviews and mental models.
Conducted holistic and ecologically valid assessment of literature review workflows. Our investigations followed a user centered approach and examined practices of 27 researchers, assessing current practices and how workflows fare under a more integrated approach to literature review with LitSense.

Problem

Researchers can have a difficult navigating through the endless amount of information and create an understanding of a field of research through literature review

Solution

Creating a web app interface to holistically support the entire literature review workflow, from finding new papers, to organizing and creating relationships, and ultimately to assisting in the comprehension of a body of literature

Methodology

1. Conducted Formative Contextual interviews with Researchers about their Literature Review Workflows and tools used during their process

2. Developed Litsense, a tool to support the holistic process and workflow of lierature review

3. Conducted usability tests with example literature review scenarios with researchers to explore effectiveness of first prototype

4. Developed second version prototype based on usability testing feedback, and conducted a follow-up one-week deployment with Litsense with 12 researchers to explore effectiveness of Litsense in ecologically valid scenario

LitSense Interface Version 1 Prototype

[Picture]

LitSense Interface Version 2 Prototype

[Picture]

LitSense components: (A) Search panel; (B) Publications list, (C) Topic Graph; (D) Paper Details tab, active when a paper node is selected; (E) Summary tab; (F) Filter by keyword; (G) Sort by criteria, with color map ; (H) Timeline view ; (I) the top menu, featuring export functions and an option to alternate the Topic Graph and Timeline views; and (J) the History view.

Resulting Publications:

Nicole Sultanum, Christine Murad, and Daniel Wigdor. 2020. Understanding and Supporting Academic Literature Review Workflows with LitSense. In Proceedings of the 2020 International Conference on Advanced Visual Interfaces (AVI '20). Association for Computing Machinery, New York, NY, USA, Article 67, 1–5. https://doi.org/10.1145/3399715.3399830