Speech and Multimodal Interfaces Laboratory

Scientists from the SPC RAS have developed the OCEAN-AI library for personality traits assessment with the aim of automating qualified personnel selection

Scientists from the SPC RAS have developed the OCEAN-AI library with the support of the research center Strong Artificial Intelligence in Industry at ITMO University. OCEANAI is an open-source library consisting of a set of algorithms for intellectual analysis of human behavior based on multimodal data for automatic personality traits (PT) assessment. The library evaluates five PT: Openness to experience, Conscientiousness, Extraversion, Agreeableness, Non-Neuroticism.


OCEAN-AI includes four main algorithms:

  1. Audio Information Analysis Algorithm (AIA).
  2. Video Information Analysis Algorithm (VIA).
  3. Text Information Analysis Algorithm (TIA).
  4. Multimodal Information Fusion Algorithm (MIF).

OCEAN-AI provides examples of solving practical tasks based on obtained PT scores:

  1. Ranking potential candidates by professional responsibilities by: professional groups and professional skills.
  2. Predicting consumer preferences for industrial goods through: the example of car characteristics and the example of mobile device application categories.
  3. Forming effective work teams: finding a suitable junior colleague and finding a suitable senior colleague.

Thus, the OCEAN-AI library provides tools to address the following practical tasks:

  1. Analysis of the personality traits of candidates and their alignment with job requirements, which significantly improves the personnel selection process and helps to make more objective and systematic decisions when ranking candidates.
  2. Analyzing the personality traits of consumers, which is useful for predicting their interests. This allows companies to better tailor their industrial goods/services to consumer preferences, making them more unique and personalized.
  3. Identifying suitable co-workers based on their personality traits, which helps build effective work teams, improve communication, and reduce team conflicts.

For more information about the library, refer to the documentation. The code is available on GitHub and GitLab.