Model of Pollution Source Information Recommendation

Environmental protection organizations and people are plagued by conflict between pollution and intrinsic features of pollution sources because it can be difficult to go through vast amount of pollution source information to discover information that is relevant to them. The primary technique for resolving this is user preference model. The UFTB algorithm is used to examine type and score data that users have viewed by creating a model of user preferences. The modified cosine similarity is generated using co-filter technique to construct recommendation model for assessing pollution source data, and default value is anticipated to optimize process. To avoid over-optimization, we should use data from user's non-preferred type, obtain optimization default prediction matrix, add similarity data to suggested formula to obtain value and use sort, locate N users with highest similarity to target user, and recommend pollution source information to target user in accordance with their preferences. Additionally, depending on a range of suggested methodologies, one may really construct a combination recommendation system.