A Conversational Recommendation Method Considering Repeated Purchase Behavior

Abstract

The invention discloses a session recommendation method considering repeated purchase behavior. This method builds a model to model the current session and the user's long-term interests, and predicts the next item the user will click. Users often have fixed behavior habits on online platforms, so this method proposes a recommendation method that considers repeated purchase behavior, which mainly consists of four parts: the first part is to obtain the user's short-term and long-term interest based on the user's current interactive session sequence. Representation and Auxiliary Behavioral (Purchasing) Representation. The second part is to choose a repeating mode or an exploratory mode based on the user's short-term and long-term interest representations. The third part is to make recommendations in the sequence of items purchased by the user in the history according to the user's short-term and long-term interest representations and auxiliary behavior (purchase) representations in the repeated mode. The fourth part is to recommend items that are not in the sequence of items purchased by the user according to the short-term and long-term interest representations of the user in the exploration mode.

Description

technical field [0001] The invention belongs to the technical field of Internet services, and in particular relates to a session recommendation method considering repeated purchase behaviors. Background technique [0002] With the development and popularity of online services, online platforms record a large amount of user behavior data. Find out the items that users are most interested in from massive data and recommend them to users, which can greatly improve user satisfaction and company revenue. At this time, the recommendation system is very important. The recommendation system can dig out the user's favorite items from a large number of items. [0003] Traditional recommendation methods mainly include three types. One is a general recommendation system, which mines the static association between users and items from the historical interaction data between users and items, including methods such as collaborative filtering and matrix decomposition. Such methods input...

AI Technical Summary

Problems solved by technology

But in the method (method), the recommended repeated items are recommended in the item sequence in the current session, and the possibility of the user repeatedly buying the same item within a short period of time, that is, within a session is very low

And this method does not take into account the user's long-term interest, which tends to be static

If the short-term interest of the user in the current session and the long-term interest of the user are highly repetitive, the probability of repeated purchase behavior is high

Images