Loacalization_and_Prediction_using_DeepLearning
localization and locatoion prediction with deep learning Since GPS signals can be unreliable this project uses wifi fingerprints in order to predict the location of users. Data: 19,937 training & 1111 validation datapoints with 529 attributes (wifi fingerprints) measured by 25 different android devices.Each WiFi fingerprint can be characterized by the detected Wireless Access Points (WAPs) and the corresponding Received Signal Strength Intensity (RSSI). The intensity values are represented as negative integer values ranging -104dBm (extremely poor signal) to 0dbM. The positive value 100 is used to denote when a WAP was not detected. During the database creation, 520 different WAPs were detected. Thus, the WiFi fingerprint is composed by 520 intensity values. Then the coordinates (latitude, longitude, floor) and Building ID are provided as the attributes to be predicted. Relative positioning and latitude and longitude prediction is the first step to modelling a succesfull location based ad campaign,followed by maping the actual area where the target(mobile devise to which the advert will be shown) then choosing the appropriate ad.
