Jan 29, 2024

Rikkyo University, AI startups embark on joint research to develop AI-driven system to give advice on job interviews



Associate Professor Tetsuo Kobayashi and Assistant Professor Sotaro Muramoto of Rikkyo University’s College of Sport and Wellness have launched a joint research project with three companies to develop a system to measure people’s positive impressions by utilizing emotion-estimation artificial intelligence (AI). The companies involved are Tripleize Co., Ltd., SHIN4NY Inc., and Exmile Co., Ltd.

1. Establishment of objective indicators which serve as basis to judge positive impressions

This impression-measurement system, driven by emotion-estimation AI, combines Tripleize’s emotion-estimation AI engine for image analysis and SHIN4NY’s emotion-estimation AI engine for audio analysis. The system enables objective measurement of the impression a person makes on other people. Kobayashi and Muramoto will conduct data analysis, the results of which will be included in a demonstration by Exmile. This demonstration will provide job-hunting advice to individuals aspiring to become flight attendants. The aim is to establish comprehensive indicators that can serve as a basis for judging people’s likability.

2. Limitations on subjective instructions in conventional job interviews

According to Mehrabian’s Law, which delineates the relative impacts of verbal, vocal, and visual cues in communication, people are more influenced by visual information (such as appearance), and vocal information (such as vocal tones), than verbal information (such as the content of people’s speech). It is thus possible to improve impressions based on facial expression and voice through training. However, such training currently centers on subjective instructions by lecturers. There is a need to establish indicators that are capable of objectively evaluating a person’s facial expression and voice. This joint research aims at judging positive impressions based on total scores on facial expressions and voices.

3. The need for AI-driven impression-measurement system is on the rise

The impression-measurement system can reduce fees for job-hunting students, such as lecturer fees, by enabling them to train themselves to improve the impression they make. Conventional instructions for job interviews are mainly conveyed verbally. This results in ineffective guidance due to the absence of objective indicators based on measuring facial expression and voice numerically. The impression-measurement system offers a solution to this problem. It is expected to address the need for dealing with such issues, which are currently observed not only in the job-hunting support industry but also in hospitality, human resource training/education, and matchmaking. Rikkyo University’s College of Sport and Wellness is actively developing indicators to enable sports instructors to conduct objective self-evaluation of their performance based on the analysis of their facial expressions. Researchers plan to provide the impression-measurement system through an app in the medium to long term, potentially growing a market for the system.


  • Tetsuo Kobayashi (Associate Professor, Department of Sport and Wellness, College of Sport and Wellness, Rikkyo University)
    We are witnessing the dawn of an era where emotions can be quantified due to advances in AI. These advances enable us to collect a new type of data that was previously unattainable. However, the raw data itself lacks inherent meaning. We need to develop methods for analyzing this data and evaluating the results of such analyses. We are dedicated to further developing analytical and evaluative methodologies to ensure that emotion data can be effectively utilized in various fields, including sports.
  • Yuichiro Yamada (President, Tripleize Co., Ltd.)
    Our company is developing systems to estimate emotions from facial expressions using an AI engine for facial authentication. We hope that the technology developed by this joint project can be used to address various issues in the communications field.
  • Luca Kudoh (CEO, SHIN4NY Inc.)
    Our company is developing systems to estimate people’s emotions based on their voice based on the slogan “making an inspiring, yet people-friendly world by using the power of data, science, and technology.” We hope this joint project will help people make a favorable emotional impression by scientifically analyzing positive impressions.
  • Yoichiro Hashimoto (President, Exmile Co., Ltd.)
    Our company offers job interview guidance for job-seeking students based on Mehrabian’s Law, focusing on enhancing visual and vocal impressions. We hope this collaborative research will establish a scientific foundation for training people in how to effectively navigate job interviews.

Profile of joint research participants

You are viewing this site in a browser that is no longer supported or secure.
For the best possible experience, we recommend that you use a modern browser.