Data - Patterns, Inference and Prediction

Context

The Internet world in general and the IoT - Internet of Things in particular has been growing in these last years and it is foreseen that it will be growing even more in the near future. Information making available by both devices and people is also growing but not always represents and add-value to the entities that really need it. Raw information has no much of an add-value since it is usually too abundant and lacking on context. A simple case of a temperature may represent a "good" or "bad" feeling depending on the humidity, rain, personal historical record, personal resilience, etc. Therefore, in order to really add value it is important to correlate, contextualize, access to historical records and then infer, predict and recommend what is more adequate to a certain circumstance. Raw data is currently already collected from different sources (sensors, gateways, social nets, TVboxes, etc.); the challenge is to process it in order to get real add-value out of it.
This project intends to identify, analyse and implement algorithms to process, correlate, infer and predict based on concrete IoT/M2M use cases.

Project objectives

This project intends to indentify, evaluate and implement mechanisms of pattern identification, inference and near real-time prediction.

Semester 1

• Analysis of the collected data (available a priori) and construct realist scenarios which can be used to validate the results.
• Study and select data analytics tools for obtaining statics and patterns.

Semester 2

• Selection and tuning of inference and prediction algorithms adopting machine learning techniques.
• Test and analysis of the results; fine tuning depending of the obtained results. Final Report.

Key skills required

• Knowledge of inference algorithms, prediction and recommendation.
• Domain in data fields analytics and machine learning.
• Knowledge in SW (Java, Linux, JBoSS, Tomcat, etc.).

Altice Labs product lines

Technology and Innovation Coordination

Want to participate in this project?

If you think you have the appropriate competences to work with us at this project, please contact us to genius@inova-ria.pt.