Item recommendation method and device

Abstract

The present invention discloses an item recommendation method and device, and relates to the field of computer technology. A specific implementation of the method includes: obtaining an intent recognition result and extracting entity information according to the conversation content input by the first user in the current conversation; judging whether the intent recognition result is the target intent; if so, generating an item push list according to the entity information; wherein the item push list includes at least one item. This implementation can solve the technical problem of poor accuracy and real-time performance.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an item recommendation method and device. Background Art

[0002] With the rapid development of the Internet, more and more services are provided and consumed through the Internet, and online customer service has also developed rapidly. With the increase in the number of online customer service users, more and more online customer service products recommend items to users through windows (such as goods, video resources, e-books, etc.), but such recommendations are often made based on user profile data.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0004] Existing recommendation solutions make recommendations based on users' offline profile data, resulting in poor accuracy and real-time performance. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides an item recommendation method and device to solve the technical problem of poor accuracy and real-time performance.

[0006] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for recommending items is provided, comprising:

[0007] Obtaining intent recognition results and extracting entity information according to the conversation content input by the first user in the current conversation;

[0008] Determining whether the intention recognition result is the target intention;

[0009] If so, an item push list is generated according to the entity information; wherein the item push list includes at least one item.

[0010] Optionally, obtaining an intent recognition result and extracting entity information according to the conversation content input by the first user in the current conversation includes:

[0011] Acquire the conversation content input by the first user in the current conversation in real time;

[0012] Using an intention recognition model to recognize the intention of the conversation content, thereby obtaining an intention recognition result; wherein the intention recognition result includes a target intention and a non-target intention;

[0013] An entity recognition model is used to extract entities of the conversation content, thereby obtaining entity information;

[0014] The intent recognition result and the entity information are stored.

[0015] Optionally, after generating the item push list according to the entity information, the method further includes:

[0016] receiving in real time the conversation content input by the first user in the current conversation;

[0017] Extracting entity information from the conversation content, and updating the stored entity information based on the currently extracted entity information;

[0018] Generate a new item push list based on the updated entity information.

[0019] Optionally, after determining whether the intention recognition result is a target intention, the method further includes:

[0020] If not, continue to obtain the intent recognition result and extract the entity information based on the conversation content input by the first user in the current conversation.

[0021] Optionally, after storing the intention recognition result and the entity information, the method further includes:

[0022] Filtering out a plurality of target entity information from each of the entity information, putting the plurality of target entity information into a first entity set, and putting the remaining entity information into a second entity set;

[0023] The first entity set is used to generate an item recommendation list.

[0024] Optionally, after generating the item push list according to the entity information, the method further includes:

[0025] The item push list is pushed to the first user.

[0026] Optionally, after generating the item push list according to the entity information, the method further includes:

[0027] Pushing the item push list to the second user;

[0028] Receive at least one target item selected by the second user in the item push list;

[0029] The at least one target item is pushed to the first user.

[0030] In addition, according to another aspect of an embodiment of the present invention, there is provided an item recommendation device, comprising:

[0031] An analysis module, configured to obtain an intent recognition result and extract entity information based on the conversation content input by the first user in the current conversation;

[0032] The recommendation module is used to determine whether the intention recognition result is the target intention; if so, generate an item push list according to the entity information; wherein the item push list includes at least one item.

[0033] Optionally, the analysis module is further used for:

[0034] Acquire the conversation content input by the first user in the current conversation in real time;

[0035] Using an intention recognition model to recognize the intention of the conversation content, thereby obtaining an intention recognition result; wherein the intention recognition result includes a target intention and a non-target intention;

[0036] An entity recognition model is used to extract entities of the conversation content, thereby obtaining entity information;

[0037] The intent recognition result and the entity information are stored.

[0038] Optionally, the recommendation module is further used to:

[0039] After generating the item push list according to the entity information, receiving in real time the conversation content input by the first user in the current conversation;

[0040] Extracting entity information from the conversation content, and updating the stored entity information based on the currently extracted entity information;

[0041] Generate a new item push list based on the updated entity information.

[0042] Optionally, the recommendation module is further used to:

[0043] After determining whether the intent recognition result is the target intent, if not, continue to obtain the intent recognition result and extract entity information based on the session content input by the first user in the current session.

[0044] Optionally, the analysis module is further used for:

[0045] After storing the intention recognition result and the entity information, filtering out a plurality of target entity information from each of the entity information, putting the plurality of target entity information into a first entity set, and putting the remaining entity information into a second entity set;

[0046] The first entity set is used to generate an item recommendation list.

[0047] Optionally, it also includes:

[0048] A push module is used to generate an item push list according to the entity information and then push the item push list to the first user.

[0049] Optionally, it also includes:

[0050] The push module is used to generate an item push list according to the entity information, and then push the item push list to the second user; receive at least one target item selected by the second user in the item push list; and push the at least one target item to the first user.

[0051] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, including:

[0052] one or more processors;

[0053] a storage device for storing one or more programs,

[0054] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the above embodiments.

[0055] According to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0056] One embodiment of the above invention has the following advantages or beneficial effects: because the intent recognition result is obtained and the entity information is extracted according to the conversation content input by the first user in the current conversation, if the intent recognition result is the target intent, a technical means is adopted to generate an item push list according to the entity information, so the technical problems of poor accuracy and real-time performance in the prior art are overcome. The embodiment of the present invention monitors the conversation content of the first user and the second user in real time, continuously analyzes the needs of the first user, and thus generates an item push list that best meets the current needs of the first user. Therefore, the embodiment of the present invention can improve the accuracy and real-time performance of item recommendations, thereby improving the user experience and facilitating users to select items.

[0057] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with specific implementation examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0059] Figure 1 is a schematic diagram of the main process of the item recommendation method according to an embodiment of the present invention;

[0060] Figure 2 is a schematic diagram of the main process of an item recommendation method according to a reference embodiment of the present invention;

[0061] Figure 3 is a schematic diagram of the main process of an item recommendation method according to another reference embodiment of the present invention;

[0062] Figure 4 is a schematic diagram of the main process of an item recommendation method according to yet another reference embodiment of the present invention;

[0063] Figure 5 is a schematic diagram of main modules of an item recommendation device according to an embodiment of the present invention;

[0064] Figure 6 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0065] Figure 7 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0067] The existing recommendation scheme is based on the user's offline profile data. This scheme has the following disadvantages:

[0068] 1) Inaccurate. Since the offline portrait depicts the status of a period of time in the past, the user may be shopping for others, so the current needs may not match the user's portrait characteristics at all. The portrait information is generated based on the user's past search, browsing, and add-to-cart behaviors, but the user may not have performed these behaviors and directly consulted customer service. Even with these behaviors, the information is still not accurate enough. For example, the user will only search for notebooks, or at most search for red notebooks, thin and light notebooks, ASUS notebooks, etc., but the user will not search for red, thin, ASUS brand, cheaper notebooks, and notebooks used for office work and watching movies instead of playing games.

[0069] 2) Not timely. It is impossible to track the changes in users’ demands during the consultation process, resulting in the recommended items not being updated according to the users’ new demands.

[0070] In order to solve the technical problems existing in the prior art, the embodiments of the present invention provide a method and device for recommending items based on the current conversation content, which solves both accuracy and real-time performance.

[0071] Figure 1 Schematic diagram of the main process of the item recommendation method according to an embodiment of the present invention. As an embodiment of the present invention, Figure 1 As shown, the item recommendation method may include:

[0072] Step 101: Obtain intent recognition results and extract entity information based on the conversation content input by the first user in the current conversation.

[0073] In an embodiment of the present invention, the conversation content of the first user (such as a consultant) is monitored in real time, and the conversation content input by the first user in the current conversation is analyzed in real time, so as to obtain the intention recognition result and entity information in real time.

[0074] Optionally, step 101 may include: acquiring the conversation content input by the first user in the current conversation in real time; using an intent recognition model to recognize the intent of the conversation content, thereby obtaining an intent recognition result; wherein the intent recognition result includes a target intent and a non-target intent; using an entity recognition model to extract the entity of the conversation content, thereby obtaining entity information; storing the intent recognition result and the entity information. The first user will clearly express his or her needs during the consultation process. For example, if he or she inquires about a notebook, it means that the user has a purchase intention and hopes to buy a notebook. If the user inquires about the price again, it represents the user's demand for the price. The user may even clearly state whether it is for office use, for playing games, or for both. Therefore, by analyzing the conversation content acquired in real time, the user's real-time intent recognition result and entity information can be obtained.

[0075] Optionally, in step 101, entry information (such as item details entry, store page entry, etc.), browsing information, add-to-cart information, link addresses entered by users, etc. may also be collected. This information may be combined with entity information to determine items to be recommended.

[0076] In the embodiment of the present invention, it is only necessary to identify the intent of the conversation content input by the user, and the identified intent results include target intent and non-target intent. Optionally, the target intent may be purchase intent, reading intent, or download intent, etc.

[0077] Optionally, the intent in the conversation content can be identified by training a binary classification model, and the model can be trained using algorithms such as linear regression, logistic regression, and deep learning (such as a support vector machine algorithm). The classification model can be used to identify whether the customer is consulting with the intention of purchasing (or reading, downloading, etc.).

[0078] Before training the classification model, the data needs to be labeled. Specifically, the historical conversation content can be manually labeled, that is, every sentence entered by the user during the conversation is labeled. The labeled labels include target intent and non-target intent.

[0079] Entity recognition requires extracting entity information from the conversation content entered by the user, such as item classification, price, size, color, author, publication, publication time, actor information, release time, etc. Specifically, item classification also includes certain commodity-specific attributes, such as mobile phones may have screen size, storage capacity, camera parameters, etc., washing machines may have motor type, energy efficiency level, with or without drying, etc., shampoo may have net content, suitable hair quality, applicable population children type, etc. Optionally, models such as CRF (conditional random field), BILSTM (bidirectional long short-term memory network) or BERT can be used to extract entity information from the conversation content. Alternatively, the BILSTM_CRF model can be used to extract entity information from the conversation content.

[0080] For example, if the user inputs the conversation content "Hello, is this blue phone out of stock?", then the intent recognition result is the target intent, and the entity information is: item-phone, color-blue. If the user inputs the conversation content "OK, send it to see", then the intent recognition result is the target intent, and the entity information is: none. If the user inputs the conversation content "I don't like it, I want a full screen, 5.5 inches or more", then the intent recognition result is the target intent, and the entity information is: screen type-full screen, size-5.5 inches. If the user inputs the conversation content "OK, I'll take a look", then the intent recognition result is the target intent, and the entity information is: none. If the user inputs the conversation content "Forget it, I won't buy it", then the intent recognition result is a non-target intent, and the entity information is: none.

[0081] Optionally, after storing the intention recognition result and the entity information, the method further includes: filtering out multiple target entity information from each of the entity information, placing the multiple target entity information into a first entity set, and placing the remaining entity information into a second entity set; wherein the first entity set is used to generate an item recommendation list. Since the entity information of the session content input by the user is extracted in real time, more and more entity information will be stored, so it is necessary to filter multiple entity information, and use the entity information that can be simultaneously input into the recommendation algorithm as the target entity information. For example, the stored entity information includes item-mobile phone, color-blue, screen type-full screen, size-5.5 inches, with or without drying-with drying, laundry specification-5kg, then "item-mobile phone, color-blue, screen type-full screen, size-5.5 inches" is placed in the first entity set, and in step 103, the recommendation algorithm is executed based on each entity information in the first entity set, thereby obtaining an item recommendation list. "With or without drying-with drying, laundry specification-5kg" is obviously irrelevant to the mobile phone, so it is placed in the second entity set.

[0082] Step 102, determine whether the intention recognition result is the target intention, if so, execute step 103.

[0083] In this step, determine whether the intent result identified in step 101 is the target intent. If so, execute step 103; if not, execute step 101, that is, continue to obtain the intent recognition result and extract the entity information according to the session content input by the first user in the current session, until the intent recognition result is the target intent, then execute step 103.

[0084] Step 103: Generate an item push list according to the entity information.

[0085] In this step, if the intent recognition result is the target intent, an item push list is generated according to the entity information extracted in step 101. The item push list includes at least one item. Specifically, the entity information can be input into a search engine, and the search engine outputs an item push list based on a recommendation algorithm.

[0086] After step 103, the method further includes: receiving the conversation content input by the first user in the current conversation in real time; extracting entity information from the conversation content, and updating the stored entity information based on the currently extracted entity information; and generating a new item push list according to the updated entity information. In the embodiment of the present invention, steps 101 and 102 are continuously executed, so it is also necessary to continuously generate an item push list according to the updated entity information to ensure real-time tracking of user needs and improve the accuracy and real-time performance of item recommendations.

[0087] Optionally, both intent recognition and entity information extraction are monitored asynchronously and continuously, and each arrival of entity information drives a recommendation process, that is, each time the entity information is updated, step 103 is triggered to update the item recommendation list to ensure that the recommendation information remains up to date.

[0088] According to the various embodiments described above, it can be seen that the present invention obtains the intent recognition result and extracts the entity information according to the conversation content input by the first user in the current conversation. If the intent recognition result is the target intent, the technical means of generating an item push list according to the entity information solves the technical problems of poor accuracy and real-time performance in the prior art. The embodiment of the present invention monitors the conversation content of the first user and the second user in real time, continuously analyzes the needs of the first user, and thus generates an item push list that best meets the current needs of the first user. Therefore, the embodiment of the present invention can improve the accuracy and real-time performance of item recommendations, thereby improving the user experience and facilitating users to select items.

[0089] Figure 2 is a schematic diagram of the main process of an item recommendation method according to a reference embodiment of the present invention. As another embodiment of the present invention, Figure 2 As shown, the item recommendation method may include:

[0090] Step 201: Acquire in real time the conversation content input by the first user in the current conversation.

[0091] Step 202: Use an intent recognition model to identify the intent of the conversation content, thereby obtaining an intent recognition result; at the same time, use an entity recognition model to extract entities of the conversation content, thereby obtaining entity information.

[0092] The conversation content of the first user (ie, the consultant) is monitored in real time, and the conversation content input by the first user in the current conversation is analyzed in real time through the intent recognition model and the entity recognition model, so as to obtain the intent recognition result and entity information respectively.

[0093] It should be pointed out that intent recognition and entity extraction can be performed simultaneously, that is, the conversation content is input into the intent recognition model and the entity recognition model respectively, and the two models output intent recognition results and entity information respectively.

[0094] Step 203: store the intention recognition result and the entity information.

[0095] Optionally, after storing the intention recognition result and the entity information, the method further includes: filtering out multiple target entity information from each of the entity information, placing the multiple target entity information into a first entity set, and placing the remaining entity information into a second entity set; wherein the first entity set is used to generate an item recommendation list. Since the entity information is extracted from the conversation content input by the user in real time, more and more entity information will be stored, so it is necessary to filter multiple entity information and use the entity information that can be simultaneously input into the recommendation algorithm as the target entity information.

[0096] Step 204, determine whether the intention recognition result is the target intention; if so, execute step 205; if not, execute step 201.

[0097] Step 205: Generate an item push list according to the entity information.

[0098] If the intent recognition result output by the intent recognition model is the target intent, the entity information output by the entity recognition model is input into the search engine, and the search engine outputs an item push list based on the recommendation algorithm.

[0099] Step 206: Push the item push list to the first user.

[0100] In this embodiment, the generated item push list may be directly pushed to the first user, and the first user may independently select items on the list.

[0101] In addition, the specific implementation content of the item recommendation method in a reference embodiment of the present invention has been described in detail in the item recommendation method described above, so the repeated content will not be described here.

[0102] Figure 3 is a schematic diagram of the main process of an item recommendation method according to another reference embodiment of the present invention. As another embodiment of the present invention, Figure 3 As shown, the item recommendation method may include:

[0103] Step 301: Acquire in real time the conversation content input by the first user in the current conversation.

[0104] Step 302: Use an intent recognition model to identify the intent of the conversation content, thereby obtaining an intent recognition result; at the same time, use an entity recognition model to extract entities of the conversation content, thereby obtaining entity information.

[0105] Step 303: store the intent recognition result and the entity information.

[0106] Step 304, determine whether the intention recognition result is the target intention; if so, execute step 305; if not, execute step 301.

[0107] Step 305: Generate an item push list according to the entity information.

[0108] Step 306: Push the item push list to the second user.

[0109] Step 307: Receive at least one target item selected by the second user in the item push list.

[0110] Step 308: Push the at least one target item to the first user.

[0111] In this embodiment, the server can directly push the generated item push list to the second user (i.e., the respondent, such as customer service). The second user selects items on the list. After the server receives the selection result returned by the second user (e.g., at least one target item in the list is selected), the server sends the selection result to the first user, and the first user can see several target items selected by the second user.

[0112] In addition, the specific implementation content of the item recommendation method in another reference embodiment of the present invention has been described in detail in the item recommendation method described above, so the repeated content will not be described here.

[0113] Figure 4 Schematic diagram of the main process of an item recommendation method according to another reference embodiment of the present invention. As another embodiment of the present invention, Figure 4 As shown, the item recommendation method may include:

[0114] Step 401: Acquire in real time the conversation content input by the first user in the current conversation.

[0115] Step 402: Use an intent recognition model to recognize the intent of the conversation content, thereby obtaining an intent recognition result, and store the intent recognition result.

[0116] Step 403: extract entities of the conversation content using an entity recognition model to obtain entity information, and store the entity information.

[0117] Step 404, determining whether the intention recognition result is the target intention.

[0118] Step 404, determine whether there is entity information.

[0119] If the intent recognition result is the target intent and has entity information, execute step 406.

[0120] Step 406: Generate an item push list based on the entity information.

[0121] Optionally, after step 406, the method further includes: receiving in real time the conversation content input by the first user in the current conversation; extracting entity information from the conversation content, and updating the stored entity information based on the currently extracted entity information; and generating a new item push list according to the updated entity information. In the embodiment of the present invention, steps 401 and 403 are continuously executed, so it is also necessary to continuously generate an item push list according to the updated entity information to ensure real-time tracking of user needs and improve the accuracy and real-time performance of item recommendations.

[0122] Optionally, both intent recognition and entity information extraction are monitored asynchronously and continuously, and each arrival of entity information drives a recommendation process, that is, each time the entity information is updated, step 103 is triggered to update the item recommendation list to ensure that the recommendation information remains up to date.

[0123] In addition, the specific implementation content of the item recommendation method in another reference embodiment of the present invention has been described in detail in the item recommendation method described above, so the repeated content will not be described here.

[0124] Figure 5 is a schematic diagram of the main modules of the item recommendation device according to an embodiment of the present invention. Figure 5 As shown, the item recommendation device 500 includes an analysis module 501 and a recommendation module 502. The analysis module 501 is used to obtain an intent recognition result and extract entity information according to the conversation content input by the first user in the current conversation; the recommendation module 502 is used to determine whether the intent recognition result is a target intent; if so, generate an item push list according to the entity information; wherein the item push list includes at least one item.

[0125] Optionally, the analysis module 501 is further used for:

[0126] Acquire the conversation content input by the first user in the current conversation in real time;

[0127] Using an intention recognition model to recognize the intention of the conversation content, thereby obtaining an intention recognition result; wherein the intention recognition result includes a target intention and a non-target intention;

[0128] An entity recognition model is used to extract entities of the conversation content, thereby obtaining entity information;

[0129] The intent recognition result and the entity information are stored.

[0130] Optionally, the recommendation module 502 is further configured to:

[0131] After generating the item push list according to the entity information, receiving in real time the conversation content input by the first user in the current conversation;

[0132] Extracting entity information from the conversation content, and updating the stored entity information based on the currently extracted entity information;

[0133] Generate a new item push list based on the updated entity information.

[0134] Optionally, the recommendation module 502 is further configured to:

[0135] After determining whether the intent recognition result is the target intent, if not, continue to obtain the intent recognition result and extract entity information based on the session content input by the first user in the current session.

[0136] Optionally, the analysis module 501 is further used for:

[0137] After storing the intention recognition result and the entity information, filtering out a plurality of target entity information from each of the entity information, putting the plurality of target entity information into a first entity set, and putting the remaining entity information into a second entity set;

[0138] The first entity set is used to generate an item recommendation list.

[0139] Optionally, it also includes:

[0140] A push module is used to generate an item push list according to the entity information and then push the item push list to the first user.

[0141] Optionally, it also includes:

[0142] The push module is used to generate an item push list according to the entity information, and then push the item push list to the second user; receive at least one target item selected by the second user in the item push list; and push the at least one target item to the first user.

[0143] According to the various embodiments described above, it can be seen that the present invention obtains the intent recognition result and extracts the entity information according to the conversation content input by the first user in the current conversation. If the intent recognition result is the target intent, the technical means of generating an item push list according to the entity information solves the technical problems of poor accuracy and real-time performance in the prior art. The embodiment of the present invention monitors the conversation content of the first user and the second user in real time, continuously analyzes the needs of the first user, and thus generates an item push list that best meets the current needs of the first user. Therefore, the embodiment of the present invention can improve the accuracy and real-time performance of item recommendations, thereby improving the user experience and facilitating users to select items.

[0144] It should be noted that the specific implementation content of the item recommendation device of the present invention has been described in detail in the above-mentioned item recommendation method, so the repeated content will not be described again here.

[0145] Figure 6 An exemplary system architecture 600 is shown to which the item recommendation method or item recommendation apparatus according to the embodiment of the present invention can be applied.

[0146] like Figure 6 As shown, system architecture 600 may include terminal devices 601, 602, 603, network 604 and server 605. Network 604 is used to provide a medium for communication links between terminal devices 601, 602, 603 and server 605. Network 604 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0147] Users can use terminal devices 601, 602, and 603 to interact with server 605 through network 604 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 601, 602, and 603, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0148] The terminal devices 601 , 602 , and 603 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0149] The server 605 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 601, 602, and 603. The backend management server may analyze and process the received data such as the item information query request, and feed back the processing results (such as target push information, item information - only an example) to the terminal device.

[0150] It should be noted that the item recommendation method provided in the embodiment of the present invention is generally executed by the server 605 , and accordingly, the item recommendation device is generally disposed in the server 605 .

[0151] It should be understood that Figure 6 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0152] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 of a terminal device suitable for implementing an embodiment of the present invention. Figure 7 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0153] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0154] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0155] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0156] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0157] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0158] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes an analysis module and a recommendation module, wherein the names of these modules do not constitute a limitation on the modules themselves in some cases.

[0159] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: obtaining an intent recognition result and extracting entity information according to the conversation content input by the first user in the current conversation; determining whether the intent recognition result is the target intent; if so, generating an item push list according to the entity information; wherein the item push list includes at least one item.

[0160] According to the technical solution of the embodiment of the present invention, the technical problem of poor accuracy and real-time performance in the prior art is overcome because the intent recognition result is obtained and the entity information is extracted according to the conversation content input by the first user in the current conversation, and if the intent recognition result is the target intent, a technical means is used to generate an item push list according to the entity information. The embodiment of the present invention monitors the conversation content of the first user and the second user in real time, continuously analyzes the needs of the first user, and thus generates an item push list that best meets the current needs of the first user. Therefore, the embodiment of the present invention can improve the accuracy and real-time performance of item recommendations, thereby improving the user experience and facilitating users to select items.

[0161] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An item recommendation method, characterized in that: include: Obtaining intent recognition results and extracting entity information according to the conversation content input by the first user in the current conversation; Determining whether the intention recognition result is the target intention; If yes, then generate an item push list according to the entity information; wherein the item push list includes at least one item; if no, then continue to obtain the intent recognition result and extract the entity information according to the conversation content input by the first user in the current conversation; After the item push list is generated according to the entity information, the method further includes: receiving in real time the conversation content input by the first user in the current conversation; Extracting entity information from the conversation content, and updating the stored entity information based on the currently extracted entity information; Generate a new item push list based on the updated entity information.

2. The method according to claim 1, characterized in that According to the conversation content input by the first user in the current conversation, an intent recognition result is obtained and entity information is extracted, including: Acquire the conversation content input by the first user in the current conversation in real time; Using an intention recognition model to recognize the intention of the conversation content, thereby obtaining an intention recognition result; wherein the intention recognition result includes a target intention and a non-target intention; An entity recognition model is used to extract entities of the conversation content, thereby obtaining entity information; The intent recognition result and the entity information are stored.

3. The method according to claim 2, characterized in that After storing the intention recognition result and the entity information, the method further includes: Filtering out a plurality of target entity information from each of the entity information, putting the plurality of target entity information into a first entity set, and putting the remaining entity information into a second entity set; The first entity set is used to generate an item recommendation list.

4. The method according to claim 1, characterized in that: After the item push list is generated according to the entity information, the method further includes: The item push list is pushed to the first user.

5. The method according to claim 1, characterized in that After the item push list is generated according to the entity information, the method further includes: Pushing the item push list to the second user; Receive at least one target item selected by the second user in the item push list; The at least one target item is pushed to the first user.

6. An item recommendation device, characterized in that: include: An analysis module, configured to obtain an intent recognition result and extract entity information based on the conversation content input by the first user in the current conversation; A recommendation module, used to determine whether the intention recognition result is the target intention; If yes, then generate an item push list according to the entity information; wherein the item push list includes at least one item; if no, then continue to obtain the intent recognition result and extract the entity information according to the conversation content input by the first user in the current conversation; The recommendation module is also used to: After generating the item push list according to the entity information, receiving in real time the conversation content input by the first user in the current conversation; Extracting entity information from the conversation content, and updating the stored entity information based on the currently extracted entity information; Generate a new item push list based on the updated entity information.

7. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

8. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images