A new recommendation strategy is proposed to enhance the location-based point of interest (POI). The POIs such as Four square, Gowall a create a curiosity to share their experience and location which they visited. This Recommendation system seems to know everything about the people's interests, passions and stays up to date with modern trends. The POIs recommendation is based on the past check-in activities of the user. The user can’t review the rating of any new cities which he/she doesn’t check-in. Even though the collaborative filtering (CF) possess numerous advantages but it suffers from some disadvantages also. It can’t predict any new (items, users), if the product or users aren’t in the training then it can’t be embedded in the query thus it creates a cold start problem. To overcome this problem, we introduced a Hybrid Co-pear Collaborative Filtering algorithm integrated into the POIs system. The combination of Cosine and Pearson similarity measures develop the multimodal hybrid method. To upgrade this system, we added a pear review analysis technique that analyses the location and provides valuable information to the new user. The simi larity measure is taking place by modifying the data into the vector. During the comparison, we can precisely detect the angle variation between any two non-zero vectors. The cold start problem is overcome and represents a better recommendation for the new users. User GPS trajectories is utilized to calculate the latitude and longitude of the users location from Geo-life dataset. The result showed a variation among the existing and proposed location recommendations and proved that the proposed system provides an optimum solution to the new users. © IJSTR 2019.