	US 12,333,589 B2
	Item recommendation method based on importance of item in conversation session and system thereof
Fei Cai, Hunan (CN); Wanyu Chen, Hunan (CN); Zhiqiang Pan, Hunan (CN); Chengyu Song, Hunan (CN); Yitong Wang, Hunan (CN); Yanxiang Ling, Hunan (CN); Xin Zhang, Hunan (CN); and Honghui Chen, Hunan (CN)
Assigned to National University of Defense Technology, Changsha (CN)
Filed by National University of Defense Technology, Hunan (CN)
Filed on Mar. 14, 2022, as Appl. No. 17/693,761.
Application 17/693,761 is a continuation in part of application No. 17/325,053, filed on May 19, 2021, abandoned.
Claims priority of application No. 202010450422.4 (CN), filed on May 25, 2020.
Prior Publication US 2022/0198546 A1, Jun. 23, 2022
Int. Cl. G06Q 30/0601 (2023.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)

CPC G06Q 30/0631 (2013.01) [G06N 5/04 (2013.01); G06N 20/00 (2019.01)]

7 Claims

1. A system for performing an item recommendation method based on importance of item in a conversation session, configured to predict an item that a user is likely to interact at a next moment from an item set as a target item to be recommended to the user, comprising:

a processor;

a non-transitory computer readable memory coupled to the processor containing program instructions for performing the item recommendation method; and

an input device;

wherein the input device is configured to record an input from the user; and

execution of the program instructions by the processor causes the processor to perform steps of an item recommendation model comprising:

obtaining an item embedding vector by embedding each item in a current conversation session to one d-dimension vector representation, and taking an item embedding vector corresponding to the last item in the current conversation session as a current interest representation of the user, wherein the current conversation session is a voice recording of the user via a microphone of a conversation conducted by the user;

obtaining an importance representation of each item according to the item embedding vector, and obtaining a long-term preference representation of the user by combining the importance representation with the item embedding vector, wherein the obtaining the importance representation of each item further comprises:

converting an item embedding vector set formed by each item embedding vector corresponding to each item in the current conversation session to a first vector space and a second vector space respectively so as to obtain a first conversion vector Q and a second conversion vector K respectively and the first conversion vector Q and the second conversion vector K are calculated according to:

where W_q∈R^d×land W_k∈R^d×lare trainable parameters corresponding to a query and a key, respectively; l is a dimension of an attention mechanism adopted while performing formulas (1) and (2); and sigmoid is a conversion function learning information from the item embedding vector in a nonlinear manner;

adopting a cross entropy function as an optimization target to learn the trainable parameters;

implementing a back propagation algorithm to train the item recommendation model;

obtaining an association matrix C between the first conversion vector Q and the second conversion vector K, wherein the association matrix C is calculated according to:

where √d is used to reduce an attention pro rata;

blocking a diagonal line of the association matrix by one blocking operation during a process of obtaining the importance representation according to the association matrix, thereby removing irrelevant items in the current conversation session;

obtaining an importance score α_iusing the association matrix C, wherein the importance score α_iis calculated according to:

and

obtaining the importance representation β_iusing a softmax layer and the importance score α_i, wherein the importance representation β_iis calculated according to:

obtaining a preference representation of the user by connecting the current interest representation and the long-term preference representation by a connection operation;

obtaining and recommending the target item to the user according to the preference representation and the item embedding vector, thus achieving the item recommendation method by focusing on importance of the item in the conversation session, wherein the target item is based on the current conversation session of the user comprises online shopping recommendations, hotel recommendations, catering recommendations, navigation recommendations;

capturing a preference of the user with unavailability of user-item interaction histories and delivering a recommended target item to the user, thereby improving accuracy of item recommendation and reducing calculation complexity of the item recommendation model; and

displaying the target item to the user via one or more output devices of the receive, wherein

the processing system is implemented with a bus architecture, represented by a bus that comprises a number of interconnecting buses and bridges;

the processing system is coupled to a transceiver that is coupled to one or more antennas; the transceiver communicates with other apparatus over a transmission medium; and transceiver receives a signal from the one or more antennas, extracts information from the received signal, and provides the extracted information to the processing system.