• Smriti Nigam

Location prediction is the key to many applications like traffic planning and controlling, weather forecasting, recom-mendation services according to the location like the hotel, food, homeland security, and travel recommendation. Previously location prediction was done with the help of history or past moving pattern of an individual but it may fail sometimes because exact prediction doesn’t reflect only past data. It divides into two phases, first phase analyses the individual preferences and the next phase is to find the social group in which an individual belongs. In this dissertation work, the focus is to solve the location prediction problem using a semi-supervised approach. In this approach, the first phase identifies the social groups of a user by using HDBSCAN clustering algorithm with a geosocial dataset which has latitude and longitude information of a user. Haversine distance is used for finding the distance between geosocial points. The next phase applies Random forest classification in identified POIs to predict the correct location of the user. It also identifies preferred POI and what are the services available on that POI. Impact of the system is to recommend location-based services to the user. Location-based services used in many ways, it can help to understand user mindset, with the help of rating and preference, it recommends the services to a user which enhances the business of the service providers. It also provide ease to the user so that they can easily access their service which the user need.

HDBSCAN clustering,Machine Learning,Haversine Distance, Geosocial data,Location Prediction,Recommendation of Services.

"Predicting next location and Recommend services using Geo-social data", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.446-451, June 2019, Available :http://www.jetir.org/papers/JETIR1907524.pdf

"Predicting next location and Recommend services using Geo-social data", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp446-451, June 2019, Available at : http://www.jetir.org/papers/JETIR1907524.pdf