INDUSTRY USES CASES
Building a Facial Expression Recognizer
Challenge:
Build a Face Expression Recognition Tool by using advanced Computer Vision algorithms. You will be able to detect Happiness, Sadness and Other emotions from Human Faces in Real time Videos
Technology Components:
Machine Vision, Micro radar, Signal, processing, ML & AI
Time:
5 Months
Technology:
6+ technologies
Ressources:
5
NewsMiner
Challenge:
NewsMiner is a software solution to crawl through selected news websites and produce a meaningful visualisation of the topics & trends by clustering similar news articles.
These topics will be characterized by relevant tags which are generated automatically from the articles. This product is in development stage.
Technology Components:
Elasticsearch, Kibana, Django, Abstractive Summary & Tag Generation
Time:
3 Months
Technology:
5+ technologies
Ressources:
5
SherDoc
Challenge:
SherDoc is a custom Document Management System that can be used to store documents and then retrieve them by using their contents powered by Machine Learning and NLP.
The AI module will create keywords from uploaded documents and help in semantic search
Technology Components:
Abstractive Summary, OCR, Django, SQL
Time:
3 Months
Technology:
5+ technologies
Ressources:
5
Fruits Recognition using CNN
Challenge:
Solve a complicated Image Classification Task with Multiple Classes. In this Problem there are more than 50 different Fruits, So, we have to Train a Image Classifier which can recognize all the distinct fruits
The AI module will create keywords from uploaded documents and help in semantic search
Technology Components:
Abstractive Summary, OCR, Django, SQL
Time:
3 Months
Technology:
4 technologies
Ressources:
5
Customers Segmentation Engine
Challenge:
Divide the customer base into several groups of individuals that share a similarity in different ways that are relevant to marketing such as gender, age, interests, and miscellaneous spending habits.
The AI module will create keywords from uploaded documents and help in semantic search
Technology Components:
Abstractive Summary, OCR, Django, SQL
Time:
3 Months
Technology:
5+ technologies
Ressources:
5
KYRIOS SYSTEM AI ANTI FRAUD IN PUBLIC TRAIN
Challenge:
A large flow of passengers: Every day, several million passengers take public transit. It is therefore very difficult for ticket inspectors to recognize fraudsters.
There is a glaring lack of information about where fraud is most prevalent, even though data exists in several of the systems. This is one of the reasons for the low effectiveness of the fight against fraud. So there are ways forward:
- The positioning of the control teams
- Screening officers in plain clothes
- Communication as a deterrent to fraud, a reminder of the rule, the standard...
- The increase in the price of fines etc...
Technology Components:
Abstractive Summary, OCR, PYTHON, SQL, Elasticsearch, Kibana, ML
Time:
3 Months
Technology:
5+ technologies
Ressources:
4
Contact Us
Enhance every aspect of operations and R&D initiatives. Together let’s see how Artificial Intelligence can help your organization.
Our address
45, Rue Wauthier 78100 Saint-Germain-en-Laye
Our e-mail address
contact@diversiti.fr
Phone
+33 7 89 67 87 91