BasisTech Chief Scientist Presents Published Work on Using NLP to Help Diagnose Mental Health Disorders

Kfir Bar delivers keynote session during Future of Text Analytics conference

Kfir Bar, Ph.D., and Chief Scientist at BasisTech, delivered the keynote address on “Adaptable NLP: A Case Study in Mental Health” during the recent Future of Text Analytics virtual summit. The presentation was based on the published work of Bar and colleagues as part of the proceedings of the Sixth Workshop on Computational Linguistics and Clinical Psychology.

The group focused on schizophrenia and the adaptability of natural language processing (NLP) software to detect the distinguishing characteristics of associated disordered speech. Their goal was to automatically and reliably detect disorganized speech patterns to help clinicians make better diagnoses and pursue appropriate treatment options.

In his presentation, Bar described the particular speech patterns, such as derailment and incoherence, that are strong indicators of thought disorders like schizophrenia. In the group’s study of Hebrew-speaking mental health patients and a control group, scores were assigned to the degree of speech derailment and coherence. Using these measures, the researchers were able to build a classification system for detecting schizophrenia through recorded interviews that achieved 81.5% accuracy.

The text analysis system they used was the Ben-Gurion University Morphological Tagger, a context-sensitive morphological analyzer for Modern Hebrew. The methodology can also be applied to models trained to detect these patterns in other languages.

“We were very encouraged by results that showed a strong degree of accuracy for predicting a schizophrenia diagnosis through text analysis. It certainly holds promise as another diagnostic tool for mental health clinicians,” said Bar

About BasisTech

Data analytics and machine learning are critical to verifying identity, understanding customers, anticipating world events, and uncovering crime. BasisTech provides businesses and governments with advanced analytics and AI-powered solutions for deriving insights from multilingual text, connecting data silos, and discovering digital evidence. Our Rosette text analytics platform employs classical machine learning and deep neural nets to extract meaningful information from unstructured data. Autopsy, our digital forensics platform, and Cyber Triage, our incident response tool, serve the needs of law enforcement, national security, and legal technologists. KonaSearch delivers deep search across Salesforce and other data sources.

Company headquarters are in Somerville, Mass., with offices in Washington, D.C., London, Tel Aviv, and Tokyo. For more information, visit