Skip to main content
We help you make the most appropriate decisions to improve environmental and health safety

Detecting health non-compliance from online comments

Due to the large number of establishments offering food services, the selection of health checks can be complex and non-exhaustive. Thanks to an AI solution for targeting and decision support, an organization called upon our teams to prioritize the investigations. Heka.ai's support has helped collect and analyze online reviews to predict the potential sanitary non-conformity of restaurants and food service establishments for better prioritization of field surveys.

  • Business/data collaboration for the most adapted and efficient AI solution possible.

  • Language understanding to improve risk detection.

  • The implication of the business in the detection of words signifying semantic non-conformities.

  • Supervised machine learning approach from an inspection history to detect relevant patterns and propose a non-conformity score.

  • Development of an interactive web application to target priority areas.

  • Thanks to the proposed prioritization, up to +100% of non-compliant establishments were checked.

Detection of pollution sources

Our teams worked with an organization whose mission is to help prevent risks that economic activities pose to the health and safety of people and property, and the environment. This institution provides public authorities, businesses, and local communities with its scientific and technical expertise to help them make the best decisions to improve environmental safety. In this context, this client wanted to develop artificial intelligence and statistical analysis tools to facilitate the detection of pollution sources.

  • Identify sources of environmental pollution with the probability of site contamination from statistical analysis.

  • Creation of an ergonomic annotation interface for validation or invalidation by experts.

  • Characterization of air pollution sources from an air sample taken every 30 minutes.

Our publications

Labeling text clusters by Heka

Labeling text clusters with keywords

We propose to explore several keyword extraction techniques to label text clusters obtained after a Text Clustering or a Topic Modeling pipeline. This work is following our previous articles about Topic Modeling and Text Clustering (here and here).

2024

Read more
image

From sensors to display, a journey towards usable…

How to obtain useful satellite images

2024

Read more
Ai Abstract Art

Generative.AI

Data Science & AI expertise combined with consulting services enable customers to embrace all aspects of Generative AI.

1 Our Generative AI Approach
2 Generative AI at a glance
3 Use cases for Generative AI
4 How should companies prepare for Generative AI adoption?

2024

Read more

Contact us

*Required fields

This site is protected by reCAPTCHA and by the Google system.

Privacy Policy and Terms of Service apply.