package
tests
docs
license
languages
mirror
acknowledgments

Expert - is a library that is designed to assess the validity of expert opinion based on dynamic intellectual analysis of video content.

"Expert" is designed to assess the validity of expert opinion based on various parameters obtained by multimodal analysis of audio, video and text data. This development can be applied in the field of education and online learning, HR processes, issues related to the analysis of information of social and political nature.

The library was developed as part of a research project funded from the centralized funds of ITMO University.

• Expert Applications
• Requirements
• Installation
• Documentation
• Publications About Expert
• Acknowledgments
• Supported by
• Contacts
• Citation

• Confidence - is a characteristic that allows to evaluate a property, the core of which is a positive assessment of an individual's own skills and abilities sufficient to achieve goals meaningful to him.
• Aggression - is a qualitative characteristic of a person's attitude toward someone or something, expressed in a state of hostility, ill-will, aggression, anger toward the object of hostility.
• Congruence - is a characteristic that allows to evaluate the level of consistency of information simultaneously transmitted by a person verbally and non-verbally (audio, video and textual channels).
• Contradiction - is a characteristic that allows to evaluate two statements of which one is the negation of the other. If two statements are in a contradictory relation, then one of them is equivalent to the negation of the other.
• Speaker Verification
• Speech Recognition and Summarization

• Python ~=3.9 (python3.9-full, python3.9-dev)
• pip >=22.0 or PDM >=2.4.8
• CUDA >= 11.7

• Case A: If You use Expert as a dependency or third-party package:

Expert can be installed with pip:

$ pip install "expert[all] @ git https://github.com/expertspec/expert.git"

or with pdm:

$ pdm add "expert[all] @ git https://github.com/expertspec/expert.git"

In the case of installation via pdm as a third-party package, after installation, run the command:

$ pdm run pip install mmcv-full~=1.7.0 --ignore-installed --no-cache-dir

This command is needed because the method of installing dependencies in pdm conflicts with the specific installation method mmcv-full.

The expert[all] entry means that dependencies from the all group will be installed. If you want to install dependencies only from a group of a certain library module, then enter the name of the required module instead of all. Installing without specifying a dependency group will result in installing a library with basic dependencies.

• Case B: If You develop and run Expert directly, install it from source:

Clone repository:

$ git clone https://github.com/expertspec/expert.git

Install all dependencies from pdm.lock file:

$ pdm sync -G all -v

or optional dependencies for each library module (check pyproject.toml):

$ pdm sync -G <group> -v

For update dependency (package) version you need change version in pyproject.toml and after execute:

$ pdm update -G <group> <package>

Run pre-commited hooks:

$ pre-commit run (all hooks only for commited changes)
$ pre-commit run --all-files (all hooks for all changes)
$ pre-commit run <hook_name> (specified hook)

Official Documentation

We acknowledge the contributors for their important impact and the participants of numerous scientific conferences and workshops for their valuable advice and suggestions.

The study is supported by the Research Center Strong Artificial Intelligence in Industry of ITMO University as part of the plan of the center's program: Development and testing of an experimental prototype of the library of strong AI algorithms in terms of the point of view of hybrid decision-making based on the interaction of AI and the decision-maker based on models of professional behavior and cognitive processes of decision-maker in difficult to formalize tasks with high uncertainty.

• Anatolii Medvedev - ML-engineer
• Ivan Smirnov - ML-engineer
• Samigulin Timur - ML-engineer
• Artyom Bondar - ML-engineer
• Alena Komarova - ML-engineer
• Andrei Laptev - Backend Developer
• Artyom Chemezov - Frontend Develop
• Olga Gofman - Scientist
• Nika Kraynovskikh - Researcher
• Anastasiya Laushkina - Project Manager, Researcher

@software{expertspec,
    title = {expert},
    author = {Laushkina, Anastasiya and Smirnov, Ivan and Medvedev, Anatolii et al.},
    year = {2023},
    url = {https://github.com/expertspec/expert},
    version = {1.0.0}
}