Recommendation algorithm combining Word2vec word vector and LSH (Local Sensitive Hash)

Abstract

The invention discloses a matrix decomposition recommendation algorithm combining a Word2Vec word vector model and LSH local sensitive hash based on cosine similarity. The method comprises the following steps: firstly, converting word similarity into vector similarity by utilizing a Word2Vec model, then performing LSH locality sensitive hash high-speed calculation based on cosine similarity to obtain a project similarity matrix, combining the obtained project similarity matrix with an original score, outputting a pre-score of an unscored project, and filling the pre-score into a training set;and finally, taking the training set as the input of an ALS matrix decomposition algorithm to obtain a recommendation result. Compared with a traditional collaborative filtering recommendation algorithm, the improved algorithm is lower in MAE value and better in performance, and meanwhile the problem that recommendation is not timely in a large amount of data can be effectively solved.

Description

technical field [0001] The present invention proposes a matrix decomposition recommendation algorithm combining Word2Vec word vector model and LSH local sensitive hash based on cosine similarity, mainly aimed at ALS. Although the matrix decomposition recommendation algorithm is better than the neighborhood-based collaborative filtering algorithm, there are still data Sparseness, accuracy, and timeliness of recommendations in a large amount of data lead to poor user experience. The method field belongs to the field of data mining based on cloud computing platform. Background technique [0002] Machine learning and data mining continue to progress in the repeated updates and improvements of the big data platform. Among them, the recommendation system is one of the representatives. The basic idea of ​​a recommendation engine is to speculate on user preferences and assist the whole process by exploring the associations between objects. As such, it is complementary to search e...

AI Technical Summary

Problems solved by technology

Such as: 1. Timeliness issues. The recommendation system needs to store relevant information of users and items. With the development of technology, the demand for recommendation systems has become larger. The information is growing rapidly, the calculation speed is very slow, the time is prolonged, and the user experience effect is not good. it is good

2. The problem of data sparsity. Most collaborative filtering recommendation algorithms use ratings to predict after similarity calculations. However, the data sparseness of the user-item rating matrix will lead to inaccurate similarity calculations, which will have serious problems in the rating prediction process. impact, will directly reduce the accuracy of the recommendation system

3. Data accuracy problem: The original data of the user-item rating matrix is ​​sparse, resulting in inaccurate similarity, which makes the recommendation system unable to give accurate recommendation results

Images