Prewedding Location Selection Recommendation System using Count Vectorization and Cosine Similarity

Abstract

Choosing the right pre-wedding location is a concern for many couples due to the many options available, which often causes confusion and is time-consuming in decision-making. Therefore, a recommendation system is needed to assist couples in determining the prewedding location that suits their preferences. This research aims to provide alternative recommendations for prewedding locations and simplify the process of selecting a suitable location. This system integrates the content-based filtering method by applying count vectorization and cosine similarity calculations to calculate and measure the level of similarity between locations based on features in the dataset when producing prewedding location recommendations. The Rapid Throwaway Prototyping method ensures the system development is done iteratively and involves direct feedback from users. The recommendation system is evaluated using the Mean Reciprocal Rank (MRR) metric to measure the effectiveness of the recommendations provided by the system. The results show that the developed prewedding location recommendation system can provide relevant location recommendations with good performance, as evidenced by the Mean Reciprocal Rank (MRR) value of 0.88, which indicates that the system is effective in placing the most relevant locations at the top of the recommendation list. The high MRR value shows the system's effectiveness in providing relevant recommendations, improving customer experience, and supporting the company's competitiveness in the prewedding documentation industry.