A Manhattan distance based hybrid recommendation system

Many online service providers use a recommendation system to assist their customers' decision-making by generating recommendations. Accordingly, this paper proposes new for tourism customers make reservations hotels with the features they need, saving time and increasing impact of personalized hotel This combined collaborative content-based filtering approaches created hybrid system. Two datasets containing customer information were analyzed Recency, Frequency, Monetary (RFM) method in order identify according purchasing nature. The main idea is establish correlations between users products decision choose most suitable product or particular user. As result exponential growth data, vast amount industry can be leveraged decision-makers decisions[20]. Filtering, prioritizing, beneficially presenting relevant reduces overload. There are following three ways that systems generate list user; content-based, collaborative-based, approaches[1]. describes each category its techniques detail. RFM Analysis used segments measuring habits. It process labeling determining values purchases ranking them on scoring model. Scoring based how recently bought (Recency), often (Frequency), purchase size (Monetary). Experimental results show accuracy behavior analysis using Manhattan distance-based greatly improved compared algorithms.