Retrieval type personalized conversation method and system

Abstract

The invention discloses a retrieval type personalized conversation method and a system applying the method through a method in the field of artificial intelligence processing. The method comprises the following steps: constructing a personalized style matching module to obtain a style matching feature vector gS (q, r, H); designing a user portrait matching module depending on the current question, and obtaining a user portrait matching vector gP (q, r, H) to measure the consistency between a user portrait and an answer candidate; and finally, combining the style matching feature vector gS (q, r, H) and the user portrait matching vector gP (q, r, H) through a fusion module to calculate a final matching score, and returning a candidate reply with the highest matching score. According to the method, modeling is carried out on personalization of the user from the conversation history of the user, the mutual relation between the speaking style of the user and the preference of the user is mined, the source of the user personalized information learned by the model can be distinguished, and finally the two parts of user personalized information are fused to sort the candidate replies, so that the most suitable personalized reply is obtained.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a search-type personalized dialog method and system. Background technique [0002] Whether in industry or academia, open-domain conversational robots have been a hot topic in the field of natural language processing in recent years. Generally speaking, there are two ways to build an open-domain dialogue robot: retrieval and generation. The former usually recalls several answer candidates through a search engine, and then selects the appropriate response from these candidates. While the latter directly utilizes models such as encoder-decoder to generate response text. Retrieval chatbots are widely used in many industrial products, such as Microsoft's Xiaoice and Ali Xiaomi. [0003] Although the current search-based open-domain dialogue robots have made many good progress, they are still constrained by some problems in the development of this industry. Among these many c...

AI Technical Summary

Problems solved by technology

[0006] 1. Although we know that there is a large amount of user personalized information in the user's dialogue history, the current methods do not directly model the user's personalized information from the user's dialogue history;

[0007] 2. Using the method based on user vectors only trains a user vector for each user. Although this vector can help the model distinguish users to a certain extent, it is difficult to explain whether it contains the user's personalized information;

[0008] 3. Using a method based on predefined personalization requires large-scale labeling of user personality data. Such data labeling requires a lot of manpower and the cost is very high

In addition, in real application scenarios, users may not be willing to fill in detailed personalized information;

[0009] 4. The user's personalized information may change over time. Whether it is based on the user vector or the method based on the predefined personalized description, it is difficult to support the update of the user's personalized information

Images