DetectBot® 

DetectBot is an Artificial Intelligence software with 12 detection features

 

Understanding videos, sounds and texts (research in knowledge bases and verbatims analysis – CareBot) but also predictive maintenance (FixBot®), fraud detection (FraudBot) and optimization of the employee life cycle (HRBot)

Tracking several people trajectories

Recognition of emotions
via the image

Recognition of objects and people via the image

Age and gender recognition via the image

Search for a person according to different criteria

Monitoring vigilance areass

Facial recognition

 Automatic number-plate recognition

Recognition of emotions via the voice

Recognition of age and gender via the voice

Identification of the person via the voice

Text searching in knowledge bases

CareBot

CareBot, the Bot, and more generally the Muvraline service, which helps you to improve the relationship between you and your consumers. We analyze the verbatims of your consumers in order, thanks to Artificial Intelligence, to evaluate them and by decision rules, to suggest you a personalized answer / action. The idea is for example to convert detractors into promoters and thus improve your Net Promoter Score℠ (NPS®).

FixBot®: predictive maintenance

FixBot®, the assistant that analyzes the logs of all the sensors present on your installations to identify the patterns that led to a failure. The idea of ​​this predictive maintenance is to implement corrective actions as soon as we detect the warning signs of a breakdown.

FraudBot: fraud detection

FraudBot, the assistant that analyzes datasets and identifies items which have an influence on the fraudulent behavior. It thus makes it possible to detect, with regard to his behavior, whether a person is likely to defraud.

HRBot: the dynamizer of human resources

HRBot, the assistant that optimizes the collaborator life cycle and its interaction with the business cycle.

Understanding videos

01. Tracking several people trajectories

The « Tracking several people trajectories » feature of DetectBot allows to do video analysis in real time, based on surveillance cameras, so you can:

  • understand visitors routes
  • analyze visitors behavior, preferences and engagement levels
  • detect abnormal situations

This « Tracking several people trajectories » feature is particularly useful in the markets below:

Transports

Identifying peak hours and traffic flow improves operational efficiency and increases customer satisfaction and usage

Security

The analysis of visitors’ routes and their behavior, centrally on all sites, allows to be alerted in real time on security threats and thus create a safe environment for visitors

Shopping centers

Understanding visitor journeys segmented by socio-demographic class and analyzing their preferences and engagement levels enables data-driven operations and sales strategies to increase revenues and improve the visitors experience

Note that this functionality can be combined with an analysis of customer catchment areas and its developments, and this from anonymized geolocation data from SFR SIM cards

Stores

Understanding the behaviors, preferences and levels of engagement of in-store visitors allows for A / B testing to identify sales, operational, and marketing strategies that are effective in increasing conversion rates and improving customers experience

02. Recognition of emotions via the image

The « Recognition of emotions via the image » feature of DetectBot can read facial expressions and understand human emotions in real time.

This « Recognition of emotions via the image » feature is particularly useful in the markets below:

Health

Quickly recognize sudden discomfort or better quantify pain in patients unable to communicate verbally

Security

Detect a lack of attention in a driver in order to increase driving safety, or automatically determine if there are unduly nervous, aggressive or fearful people in public places

Marketing

Assess, in a non-invasive and reliable way, the attractiveness of products as well as the capacity of advertising content to effectively attract the attention and interest of the viewer

Transports

Assess the stress level of passengers landing or arriving at the station or detecting suspicious behavior

03. Recognition of objects and people via the image

The « Recognition of objects and people via the image » feature of DetectBot  identifies, from surveillance cameras, objects and people, specifying, for each item, the percentage of accuracy of the algorithm.

This « Recognition of objects and people via the image » feature is particularly useful in the markets below:

Analysis of television audiences

Security

Targeted dynamic display

Search for an object or a person

04. Age and gender recognition via the image

The « Age and gender recognition via the image » feature of DetectBot can identify the age and gender of one or more person(s).

This « Age and gender recognition via the image » feature is particularly useful in the markets below:

Analysis of television and digital signage audiences

Security

Targeted ads

Looking for a person

05. Search for a person according to different criteria

 

 

The « Search for a person according to different criteria » feature of DetectBot can search a person according to the criteria below:

  • Colors and types of clothing
  • Age and gender
  • With features such as:
    1. Wearing bag or backpack
    2. Wearing hat
    3. Wearing glasses

06. Monitoring vigilance areas

 

The « Monitoring vigilance areas » feature of DetectBot can trigger an alarm when one or more person(s) approach(es) the vigilance area you set.

This « Monitoring vigilance areas »  feature is particularly useful in the markets below:

Banks

Security

Industrial sites

Shops

07. Facial recognition

The « Facial recognition » feature of DetectBot can identify a person from his picture.

This « Facial recognition » feature is particularly useful in the following markets:

Automation

Security

Biometrics

Indexing images

08. Automatic number-plate recognition

The « Automatic number-plate recognition » feature of DetectBot is a solution for recognizing alphanumeric license plates of static or moving vehicles, from CCTV cameras or photos.

This « Automatic number-plate recognition » feature is particularly useful in the following markets:

Access control

to manage entry / exit authorizations on an industrial site or a company

Parking management

in a car park, a car fleet, an urban parking area, etc.

Traceability

of vehicles entering and leaving an industrial site or a production unit

Search for a vehicle

according to its license plate

Understanding sound

09. Recognition of emotions via the voice

The « Recognition of emotions via the voice » feature of DetectBot makes it possible to detect in real time which emotion(s) arise(s) from the human voice.

This « Recognition of emotions via the voice » feature is particularly useful in the markets below:

Health

Estimate the urgency of calls made to hospitals in order to prioritize them

Call centers

Evaluate client satisfaction while waiting to determine the order of care, distinguish deceptive speech from truthful words

Marketing

Suggest specific content or services based on the user’s mood

Transport

Suggest to the bus driver to stop the vehicle if high anger or stress in the users’ voice is detected

10. Recognition of age and gender via the voice

The « Recognition of age and gender via the voice » feature of DetectBot identifies whether it is an adult or a child, a man or a woman.

This « Recognition of age and gender via the voice » feature is particularly useful in the markets below:

Radio audience analysis

Security

Targeted radio ads

Looking for a person

11. Identification of the person via the voice

The « Identification of the person via the voice » feature of DetectBot identifies a person based on his/her voice.

The « Identification of the person via the voice » feature is particularly useful for customer services handling confidential data, in particular in the following sectors: banking, insurance, administration.

Understanding of texts

12. Text searching in knowledge bases

It is a research tool based on machine learning algorithms adaptive to the corpus and to customer needs (Chatbot customer service, research assistance tool in knowledge bases, etc).