We live in a society in which smartphones have become undoubtedly an extremity of our body, these cell phones collect a great amount of data that can be taken advantage of. Nowadays, we are aware that companies are able to predict our tastes, analyzing the navigation data of our smartphone (history, likes in social media, etc.) and offer customized advertising based on this; but the society does not think it could be possible to estimate our personality through data collected by the smartphone.
The main goal of this work is to design and implement a system that is able to estimate our type of personality from certain data. This data is obtained through the sensors of the smartphone (proximity, location, accelerometer, etc.), as well as our phone communication activity (SMSs, voice calls, ...).
If this prediction is made correctly, the data collected by the phones could be especially valuable for the companies and would be a great alternative to some traditional methods like surveys that cost much more.
All this is possible thanks to the advances in machine learning, giving rise to other studies that relate the mood of a person with the way in which the timeline of social networks is consulted.
The tasks to be performed are divided into the following phases:
Phase 1: Review of literature about the topic.
Phase 2: Learning techniques and tools for developing the work.
Phase 3: Analysis of the data obtained for detecting patterns that allow us to find the difference between some personality types or others.
Phase 4: Study and search of automatic learning algorithms that best fit this case.
Phase 5: Software development and experimentation.
Phase 6: Results analysis for obtaining resemblance with reality.
Phase 7: Writing the final book.