NATURAL SEMANTIC INFORMATION-BASED EYE TRACKING USING IMPROVED NEEDLEMAN–WUNSCH AND SUBSMATCH METHODS

Xiaowei Wang,Lubiao Zhou,Jinke Wang,Shinichi Tamura
DOI: https://doi.org/10.1142/s0219519422400140
2022-01-01
Journal of Mechanics in Medicine and Biology
Abstract:This paper proposed an improved natural semantic information-based eye-tracking method combing Needleman–Wunsch and SubsMatch technologies in psychological assessments. The natural semantic information of the self-assessment scale in psychological measurement was combined with the participants’ eye-tracking data, and a time-space similarity method was proposed to calculate the eye-tracking differences between different participants in the questionnaire response and carry out visual pattern recognition to achieve the screening of the target population in psychological measurement. The proposed method was evaluated by screening a sample at high risk of depression. The comparative results showed that the average screening accuracy of the sample under the stimulation of a single topic was 80.13% and increased to 97.37% after a dimensionality reduction. This paper verified the objectivity and effectiveness of the semantic information-based eye-tracking time-space similarity method and proved it to be a promising objective tool to assist psychological measurement.
What problem does this paper attempt to address?