Information search method and device, electronic equipment and storage medium

Abstract

The invention provides an information search method and device, electronic equipment and a storage medium, belongs to the technical field of data processing, and particularly relates to the technical field of intelligent search. According to the specific implementation scheme applied to the client side, in response to a first search text input by a user, the first search text is sent to a server side; obtaining information of a plurality of first intelligent agents matched with the first search text and question and answer results generated by the first intelligent agents by taking the first search text as a question, wherein the information is fed back by the server side; and taking the information of each first agent and the question and answer result generated by each first agent as a search result, and displaying the search result in a display interface. By applying the information search method provided by the invention, the question and answer result can be displayed to the user more intuitively.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, in particular to the field of intelligent search technology, and more particularly to an information search method, device, electronic device and storage medium. Background Art

[0002] When using the rich functions provided by the client, users may need to search for various information they need. To facilitate users in searching for information, a search box is usually provided in the client. Users can enter search text in the search box, and the client searches for information based on the search text, thereby providing search results to users. Summary of the invention

[0003] The present disclosure provides an information search method, device, electronic device and storage medium.

[0004] In a first aspect, an embodiment of the present disclosure provides an information search method, which is applied to a client and includes:

[0005] In response to a user inputting a first search text, sending the first search text to a server;

[0006] Obtaining information of a plurality of first agents matched with the first search text and question-answer results generated by each first agent using the first search text as a question fed back by the server;

[0007] The information of each first intelligent agent and the question and answer results generated by each first intelligent agent are used as search results, and the search results are displayed in a display interface.

[0008] In a second aspect, the present disclosure provides an information search method, which is applied to a server, including:

[0009] Get the first search text sent by the client;

[0010] Obtaining the matching degree between the first search text and each existing agent;

[0011] In the order of matching degree from high to low, recall the second number of agents as the first agent;

[0012] Obtaining question-and-answer results generated by each first agent using the first search text as a question;

[0013] Feedback the information of each first agent and the question and answer results generated by each first agent to the client.

[0014] In a third aspect, an embodiment of the present disclosure provides an information search device, applied to a client, comprising:

[0015] A sending module, configured to send a first search text to a server in response to a user inputting the first search text;

[0016] A first acquisition module, used to obtain information of multiple first agents matched with the first search text and question-answer results generated by each first agent using the first search text as a question fed back by the server;

[0017] A display module is used to use the information of each first intelligent agent and the question and answer results generated by each first intelligent agent as search results, and display the search results in a display interface.

[0018] In a fourth aspect, an embodiment of the present disclosure provides an information search device, which is applied to a server, and includes:

[0019] A first acquisition module, used to acquire a first search text sent by a client;

[0020] A second obtaining module, used to obtain the matching degree between the first search text and each existing intelligent agent;

[0021] A recall module, used for recalling the second number of agents as the first agent in the order of matching degree from high to low;

[0022] A second acquisition module is used to obtain the question and answer results generated by each first agent using the first search text as the question;

[0023] A feedback module is used to feed back information of each first agent and question-answering results generated by each first agent to the client.

[0024] In a fifth aspect, an embodiment of the present disclosure provides an electronic device, including:

[0025] at least one processor; and

[0026] a memory communicatively connected to the at least one processor; wherein,

[0027] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any method described in the first aspect or the second aspect above.

[0028] In a sixth aspect, an embodiment of the present disclosure provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute any of the methods described in the first or second aspects above.

[0029] In a seventh aspect, an embodiment of the present disclosure provides a computer program product, including a computer program, wherein the computer program implements any of the methods described in the first aspect or the second aspect when executed by a processor.

[0030] The solution provided by the embodiment of the present disclosure displays information of multiple agents and question-and-answer results generated by each first agent using the first search text as a question to the user according to the first search text input by the user, and can intuitively display the question-and-answer results to the user. Moreover, the question-and-answer results are generated by agents in a specific field, which attracts the user to further communicate with the agents, thereby satisfying the user's further search demands in a specific field.

[0031] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0033] Figure 1 This is a first flow chart of an information search method applied to a client provided by an embodiment of the present disclosure;

[0034] Figure 2 is a schematic diagram of a display interface of a network disk client provided by an embodiment of the present disclosure;

[0035] Figure 3 is a schematic diagram of an example of a display method of search results provided by an embodiment of the present disclosure;

[0036] Figure 4 It is a flowchart of a first example of a hidden information expansion process provided by an embodiment of the present disclosure;

[0037] Figure 5 This is a second flow chart of the information search method applied to a client provided by an embodiment of the present disclosure;

[0038] Figure 6 is a flowchart of a second example of a hidden information expansion process provided by an embodiment of the present disclosure;

[0039] Figure 7 It is a flowchart of a hidden information expansion method provided by an embodiment of the present disclosure;

[0040] Figure 8 This is a first flow chart of the information search method applied to the server provided by the embodiment of the present disclosure;

[0041] Fig. 9is a second flow chart of the information search method applied to the server provided by the embodiment of the present disclosure;

[0042] Fig.10 is a flowchart of a method for obtaining a first matching degree provided in an embodiment of the present disclosure;

[0043] Fig.11 is a structural schematic diagram of an information search device applied to a client provided by an embodiment of the present disclosure;

[0044] Fig.12 It is a structural diagram of an information search device applied to a server provided by an embodiment of the present disclosure;

[0045] Fig.13 It is a structural schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0046] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure 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 disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0047] When using the rich functions provided by the client, users may need to search for various information they need. To facilitate users in searching for information, a search box is usually provided in the client. Users can enter search text in the search box, and the client searches for information based on the search text, thereby providing search results to users.

[0048] In order to respond to a user's information search request, the present disclosure provides an information search method, which is described in detail below through specific embodiments.

[0049] See also Figure 1 , provides an information search method applied to the client, such as Figure 1 As shown, the method includes the following steps S101-S103.

[0050] The above client can be deployed on electronic devices such as mobile phones, tablet computers, and computers.

[0051] Step S101: in response to a user inputting a first search text, sending the first search text to a server.

[0052] The first search text is the text input by the user in the search area displayed on the display interface corresponding to the client.

[0053] The client can detect whether the user enters text in the search area displayed on the display interface corresponding to the client. If it is detected that the user enters text in the search area, the client can obtain the text entered by the user in the search area, that is, the first search text.

[0054] For example, Figure 2 In the display interface of the network disk client shown in the figure, the user enters the text "How is the weather today?" in the search box displayed at the top of the display interface and clicks the search button on the right. After the network disk client detects that the user clicks the search button, it obtains the first search text "How is the weather today?" entered by the user in the search box.

[0055] After obtaining the first search text input by the user, the client may send the first search text to the corresponding server. The server may obtain the first search text sent by the client.

[0056] Step S102, obtaining information of multiple first agents matching the first search text and question-answering results generated by each first agent using the first search text as a question from the server.

[0057] The intelligent agent mentioned in the present disclosure refers to a computer program based on a large language model that can autonomously complete a given task. The above-mentioned intelligent agent can interact with the user through natural language, that is, based on its own knowledge base in a specific field, it uses the large prediction model to generate question-answering results based on the text input by the user.

[0058] For example, the above-mentioned intelligent agent may involve specific fields such as law, study, postgraduate entrance examination or medicine, in which the intelligent agent can interact with users well and answer users' questions.

[0059] The agent information may include at least one of the following information: name, icon, and introduction.

[0060] The server can match the first search text with the stored agent information, and feedback the information of multiple first agents and the question and answer results generated by each first agent using the first search text as a question to the client according to the matching results. The agent information fed back by the server can include the name and icon of the agent.

[0061] The client can obtain information of multiple first agents fed back by the server and question-and-answer results generated by each first agent using the first search text as a question.

[0062] Step S103, using the information of each first intelligent agent and the question and answer results generated by each first intelligent agent as search results, and displaying the search results in a display interface.

[0063] After acquiring the information of the first intelligent agent and the question and answer results generated by each first intelligent agent using the first search text as the question, the client can display the information of the first intelligent agent and the corresponding question and answer results on the display interface.

[0064] In a possible implementation, the client may display an identifier of each first agent, such as a name and / or icon, and display the question-and-answer results generated by the first agent using the first search text as the question below the identifier.

[0065] For example, the client receives information about 10 first agents sent by the server, and question-answer results generated by the 10 first agents using the first search text as a question. The 10 first agents are agents 1-10.

[0066] like Figure 3 As shown, the client can display the name of agent 1, the question and answer results generated by agent 1 with the first search text as the question, the name of agent 2, the question and answer results generated by agent 2 with the first search text as the question, the name of agent 3, the question and answer results generated by agent 3 with the first search text as the question, the name of agent 4, the question and answer results generated by agent 4 with the first search text as the question, and so on in the display interface from top to bottom.

[0067] The solution provided by the embodiment of the present disclosure displays information of multiple agents and question-and-answer results generated by each first agent using the first search text as a question to the user according to the first search text input by the user, and can intuitively display the question-and-answer results to the user. Moreover, the question-and-answer results are generated by agents in a specific field, which attracts the user to further communicate with the agents, thereby satisfying the user's further search demands in a specific field.

[0068] In a possible implementation, the search results may also be displayed in the following manner, that is, the above step S103 may be implemented in the following manner: displaying the information of the second agent and the question and answer results generated by the second agent in the display interface, and hiding the information of the third agent and the question and answer results generated by the third agent in the display interface.

[0069] Among them, the second agent includes: the first number of agents among multiple first agents in order of matching degree with the first search text from high to low, and the third agent includes: the agents among multiple first agents except the second agent.

[0070] When displaying search results, in addition to displaying the information of the second agent and the corresponding question and answer results, the client can display a hidden information expansion button on the display interface to prompt the user that there is hidden content. The hidden information expansion button can be displayed as words such as "expand" or "see more".

[0071] Continuing with the above example, the client receives 10 first agents 1-10 and the corresponding question-and-answer results sent by the server. The order of the matching degree of these 10 first agents from high to low is agent 1, agent 2, agent 3, ..., and the first number is 2. The client can use the first two agents 1 and 2 in the above order as the second agent, and the remaining agents 3-10 as the third agent.

[0072] like Figure 4 As shown in the display interface on the far left, the client can display the name of agent 1, the question and answer results generated by agent 1 with the first search text as the question, the name of agent 2, the question and answer results generated by agent 2 with the first search text as the question, and display an "expand" button below the question and answer results generated by agent 2 with the first search text as the question, that is, hide the information expansion button.

[0073] In this way, since the second agent has a higher degree of match with the first search text searched by the user, it can answer the user's search intention more accurately. The information of the third agent with a lower ranking and the question and answer results generated by the third agent are hidden on the display interface to highlight the second agent with a higher matching degree and the question and answer results generated by the second agent, which helps users to obtain more accurate question and answer results generated by the second agent more quickly.

[0074] After the client hides and displays the information of the third agent and the question and answer results generated by the third agent in the above manner, the user can perform a hidden information expansion operation to make the client display the above hidden content on the display interface. In response to the hidden information expansion operation, the client displays the information of the third agent and the question and answer results generated by the third agent in the display interface.

[0075] The hidden information expansion operation may be a click operation of clicking a hidden information expansion button on the display interface. The client may detect the click operation of the area where the hidden information expansion button is located on the display interface to detect the hidden information expansion operation.

[0076] like Figure 4 As shown in the display interface in the middle, the user clicks the "Expand" button, and the client detects the click operation in the area where the "Expand" button is located, that is, detects the hidden information expansion operation.

[0077] The client can then display the third agent information and the question and answer results generated by the third agent on the display interface. The information displayed on the expanded display interface is as follows: Figure 4 The display interface on the far right is shown.

[0078] In this way, when the user is not satisfied with the information of the second agent and the question and answer results generated by the second agent, the client can respond to the user's hidden information and carry out operations, and further display the question and answer results generated by the third agent to the user to meet the user's desire to obtain further information.

[0079] In a possible implementation, after sending the first search text to the server, in addition to the information of the first agent and the corresponding question and answer results, the client can also receive the information of the fourth agent fed back by the server. Specifically, the present disclosure also provides the following information search method, such as Figure 5 As shown, the method includes the following steps S501-S504.

[0080] Step S501: in response to a user inputting a first search text, sending the first search text to a server.

[0081] Step S502: obtaining information of multiple first agents matching the first search text and question-answering results generated by each first agent using the first search text as a question from the server.

[0082] The above steps S501-S502 are respectively Figure 1 In the illustrated embodiment, steps S101-S102 are the same and will not be described in detail here.

[0083] Step S503, obtaining information of multiple fourth agents matching the first search text fed back by the server.

[0084] The matching degree between the fourth agent and the first search text is less than the matching degree between the first agent and the first search text.

[0085] The server can send the information of the first agent and the corresponding question and answer result and the information of the fourth agent to the client at the same time, or can send them one after another, which is not limited. The client receives the above information and question and answer result sent by the server.

[0086] Step S504, displaying the information of the first agent and the question and answer results generated by the first agent in the display interface, and hiding the information of the fourth agent in the display interface.

[0087] The client may display the information of the first agent and the question and answer results generated by the first agent on the display interface in the same manner as in the above step S103, and hide the information of the fourth agent in the display interface.

[0088] When displaying search results, in addition to displaying the information of the first intelligent agent and the corresponding question and answer results, the client can display a hidden information expansion button on the display interface to prompt the user that there is hidden content.

[0089] For example, the client receives the information of the first agent 1-3 and the corresponding question-and-answer results sent by the server, and the client also receives the information of the fourth agent 4-10 sent by the server.

[0090] The client can follow the above analogy Figure 4 The leftmost display interface displays the information of the first agent 1-3 and the corresponding question and answer results in a manner as shown, and displays an "expand" button below all the above displayed contents. Specifically, Figure 6 The client can then display the information of the fourth agent below the above content in response to the hidden information expansion operation. Figure 4 The expansion process shown in will not be repeated here.

[0091] In this way, for the fourth agent whose question and answer results are not sent by the server, the client hides the information displayed by the fourth agent, making the display interface more concise, thereby highlighting the information of the first agent, which helps users to obtain the question and answer results generated by the first agent more quickly.

[0092] After hiding and displaying the information of the fourth agent in the above manner, the user can perform a hidden information expansion operation, so that the client can display the above hidden content on the display interface. In addition to displaying the above hidden information of the fourth agent, the client can also request the server to generate an answer to the question using the first search text as the fourth agent. Accordingly, the present disclosure also provides the following hidden information expansion method, such as Figure 7 As shown, the method includes the following steps S701-703.

[0093] Step S701, in response to the hidden information expansion operation, requests the server for the question and answer result for the fourth agent.

[0094] The user can click on the area where the hidden information expansion button is located on the display interface, such as Figure 6 The second display interface from the left in the middle is shown to obtain the relevant information of the hidden fourth agent. The client can detect the click operation in the area where the hidden information expansion button is located on the display interface to detect the hidden information expansion operation. After detecting the hidden information expansion operation, the client sends a question and answer result request to the server, and the request can carry information such as the identifier of the fourth agent to request the server for the question and answer result generated by the fourth agent with the first search text as the question.

[0095] Step S702, receiving the question and answer results generated by the fourth agent using the first search text as the question and received from the server.

[0096] After receiving the request from the client, the server can feedback to the client the question and answer results generated by the fourth agent using the first search text as the question, and the client receives the question and answer results generated by the fourth agent using the first search text as the question from the server.

[0097] Step S703, displaying the information of the fourth agent and the question and answer results generated by the fourth agent in the display interface.

[0098] The client can display the information of the fourth agent and the question and answer results generated by the fourth agent below the information of the first agent and the corresponding question and answer results in the display interface, such as Figure 6 As shown in the rightmost display interface.

[0099] In addition, in the process of the client sending a request to the server, if Figure 6 As shown in the second display interface from the middle right, the client can also display prompts such as "Loading..." in the display interface to prompt the user that information related to the intelligent agent is being loaded.

[0100] In this way, when the user is not satisfied with the information of the first agent and the question and answer results generated by the first agent, the client can respond to the user's hidden information to carry out operations, further request the fourth agent to generate question and answer results with the first search text as the question from the server, and display the obtained question and answer results generated by the fourth agent to the user to meet the user's demand for further information.

[0101] Corresponding to the above-mentioned information search method applied to the client, the present disclosure also provides an information search method applied to the server, such as Figure 8 As shown, the method includes the following steps S801-S805.

[0102] Step S801: Acquire the first search text sent by the client.

[0103] The client obtains the first search text input by the user in the search area, and then sends the first search text to the server. The server can obtain the first search text sent by the client.

[0104] Step S802, obtaining the matching degree between the first search text and each existing intelligent agent.

[0105] The server can store information about agents created by different users. The owners of these agents can be the server or the user, and the user can authorize the server to use these agents.

[0106] The server can perform text matching between the first search text and the information of each existing intelligent agent to obtain the matching degree between the first search text and each existing intelligent agent.

[0107] Step S803, recalling the first second number of agents in descending order of matching degree as the first agent.

[0108] The server can sort the existing agents according to the matching degree obtained above, and recall the first second number of agents as the first agent.

[0109] Step S804, obtaining the question and answer results generated by each first agent using the first search text as the question.

[0110] The server can input the first search text into the first agent, and then run the first agent to obtain the question and answer results generated by the first agent using the first search text as the question.

[0111] Step S805: Feedback the information of each first agent and the question and answer results generated by each first agent to the client.

[0112] The client may feed back the information of each first agent and the question-and-answer results generated by each first agent to the client according to the ranking obtained according to the matching degree.

[0113] The solution provided by the embodiment of the present disclosure displays information of multiple agents and question-and-answer results generated by each first agent using the first search text as a question to the user according to the first search text input by the user, and can intuitively display the question-and-answer results to the user. Moreover, the question-and-answer results are generated by agents in a specific field, which attracts the user to further communicate with the agents, thereby satisfying the user's further search demands in a specific field.

[0114] In a possible implementation manner, the present disclosure further provides an information search method applied to a server, such as Fig. 9 As shown, the method includes the following steps S901-S907.

[0115] Step S901: obtaining a first search text sent by the client.

[0116] Step S902, obtaining the matching degree between the first search text and each existing intelligent agent.

[0117] Step S903, recalling the first second number of agents in descending order of matching degree as the first agent.

[0118] Step S904, obtaining the question and answer results generated by each first agent using the first search text as the question.

[0119] Step S905: Feedback the information of each first agent and the question and answer results generated by each first agent to the client.

[0120] The above steps S901-S905 are respectively Figure 8Steps S801-S805 in the illustrated embodiment are the same and will not be described in detail here.

[0121] Step S906, recalling the first third number of agents from the existing agents except the first agent in descending order of matching degree as the fourth agent.

[0122] After recalling the first second number of agents, the server can also recall a third number of agents from the remaining agents in order of matching degree from high to low as the fourth agent.

[0123] Step S907, feeding back information of each fourth agent to the client.

[0124] For the fourth agent, the server may not need to obtain the question and answer results generated by the fourth agent using the first search text as the question, but may directly feedback the information of the fourth agent to the client.

[0125] The above steps S906-S907 and steps S804-S805 are executed in no particular order and can be executed in parallel.

[0126] In this way, only the question and answer results generated by the second number of first agents need to be fed back to the client, and for the fourth agent with a lower matching degree than the first agent, only information is fed back to the client. This can reduce the required computing resources, reduce the waiting time of the client, and thus improve user satisfaction.

[0127] In a possible implementation, the degree of match between the first search text and each existing intelligent agent can also be obtained in the following manner, that is, the above step S802 can be implemented according to the following steps: for each existing intelligent agent, obtain the first degree of match and / or the second degree of match between the first search text and the existing intelligent agent, and determine the degree of match between the first search text and the existing intelligent agent based on the obtained degree of match.

[0128] The first matching degree is the matching degree between the first search text and the description information of the existing intelligent agent, and the second matching degree is the matching degree between the field to which the first search text belongs and the field to which the existing intelligent agent belongs.

[0129] The server can use any text matching model or algorithm to match the first search text with the description information of each existing agent to obtain a first matching degree between the first search text and each existing agent. The above text matching model or algorithm can be a bag of words model, a vector space model or a similarity algorithm.

[0130] The server can also use any semantic analysis model or algorithm to perform semantic analysis on the first search text, determine the field to which the first search text belongs, and then match the field to which the first search text belongs with the field to which the existing intelligent agent belongs to obtain a second matching degree. The above semantic analysis model or algorithm can be a naive Bayes algorithm, a support vector machine algorithm, or a graph neural network model.

[0131] After obtaining the first matching degree and the second matching degree, the server may use the first matching degree or the second matching degree as the matching degree between the first search text and the existing intelligent agent, or use the sum of the first matching degree and the second matching degree as the matching degree between the first search text and the existing intelligent agent.

[0132] In this way, the matching degree between the first search text and the existing intelligent agent obtained by combining the first matching degree and the second matching degree is more accurate, and the intelligent agent with a higher matching degree with the first search text can be obtained more accurately, and then the intelligent agent and the corresponding question and answer results can be fed back to the client more accurately.

[0133] In one possible implementation, the server may determine the degree of match between the first search text and the existing agent in the following manner: if the obtained degree of match includes a first degree of match and a second degree of match, then based on the weight coefficients corresponding to the first degree of match and the second degree of match, respectively, a weighted sum calculation is performed on the first degree of match and the second degree of match, and the weighted sum calculation result is used as the degree of match between the first search text and the existing agent.

[0134] The above weight coefficients can be adjusted according to actual conditions.

[0135] In this way, the matching degree between the first search text and the existing intelligent agent obtained by weighted summing of the first matching degree and the second matching degree is more accurate, and the intelligent agent with a higher matching degree with the first search text can be obtained more accurately, and then the intelligent agent and the corresponding question and answer results can be fed back to the client more accurately.

[0136] The disclosed embodiment also provides a method for obtaining a first matching degree, such as Fig.10 As shown, the server may adopt the following steps S1001-S1002 to obtain a first matching degree.

[0137] Before obtaining the first matching degree between the first search text and each existing intelligent agent in the following manner, the server may first perform keyword segmentation on the first search text. Specifically, the server may use any word segmentation model or algorithm to perform keyword segmentation, such as a forward maximum matching algorithm, an n-gram language model, or a hidden Markov model.

[0138] Step S1001, matching the keywords in the first search text with various types of description information of existing intelligent agents to obtain the sub-matching degree corresponding to each type of description information.

[0139] The server uses any text matching model or algorithm to perform text matching on the keywords in the first search text with the description information of each type of existing intelligent agent, and obtains the sub-matching degree between the first search text and each type of description information, that is, the sub-matching degree corresponding to each type of description information. The above text matching model or algorithm can be a bag of words model, a vector space model or a similarity algorithm.

[0140] In a possible implementation, each type of description information includes at least one of the following information: the name, introduction, recommendation words, and agent knowledge base of an existing agent.

[0141] Step S1002 : Based on the weight coefficients corresponding to the respective types of description information, weighted sum calculation is performed on the sub-matching degrees corresponding to the respective types of description information, and the weighted sum calculation result is used as the first matching degree.

[0142] After obtaining the sub-matching degree corresponding to each type of description information, for each agent, the server can multiply the sub-matching degree corresponding to each type of description information of the agent by the corresponding weight coefficient respectively, and then add these products to obtain the first matching degree of the agent.

[0143] The above weight coefficients can be adjusted according to actual conditions.

[0144] In this way, the sub-matching degree is calculated for each type of description information of the agent, and then the first matching degree is obtained by weighted summation based on the sub-matching degree, so that the first matching degree obtained in this way is more accurate. The matching degree between the first search text and the existing agent obtained based on the first matching degree and the second matching degree is also more accurate, and the agent with a high matching degree with the first search text can be obtained more accurately, and then the agent and the corresponding question and answer results can be fed back to the client more accurately.

[0145] In a possible implementation, the server may also determine whether the first search text satisfies the conditions for recalling the agent to respond to the user search in the above manner, and recall the agent if the conditions are met. Specifically, after obtaining the first search text sent by the client in step S801, the server may determine whether the first search text satisfies at least one of the following conditions 1-3. If so, step S802 is executed to obtain the matching degree between the first search text and each existing agent.

[0146] Condition 1: The first search text indicates searching for resources provided by non-subject functions provided by the client.

[0147] The client can provide users with resources provided by the main function and resources provided by non-main function.

[0148] For resources provided by the subject function, the server corresponding to the client may store the resources provided by the subject function. If the first search text indicates to search for resources provided by the subject function provided by the client, the server may directly recall these stored resources and send them to the client, and the client displays these resources to the user without recalling the agent.

[0149] For resources provided by non-subject functions, the server corresponding to the client does not store the resources provided by non-subject functions. If the first search text indicates to search for resources provided by non-subject functions provided by the client, this search demand is also called a long-tail demand. The server can execute the subsequent step S802, recall the intelligent agent and then send it to the client, and the client displays the intelligent agent related information to the user to meet the user's long-tail demand for resources provided by non-subject functions.

[0150] For example, the main function of the network disk client is to store files. For files or other public files stored by users in the network disk, these are resources provided by the main function of the network disk. If the user enters a first search text to indicate a search for a file stored in the network disk, after receiving the first search text sent by the client, the server can match the first search text with the stored files, and then recall the files matching the first search text and send them to the client, which then displays the files to the user.

[0151] However, if the user enters a first search text such as a knowledge question such as "What's the weather like today?", the server usually does not store the corresponding file. For the above long-tail demand, the server cannot find a matching file when matching the first search text with the stored files according to the above process. The server can execute the subsequent step S802 to recall the agent and then send it to the client, and the client displays the agent-related information to the user to meet the above long-tail demand of the user.

[0152] When the first search text satisfies condition 1, that is, when the user's search demand is for resources provided by non-subject functions provided by the search client, the subsequent step S802 is executed to recall the agent to provide the user with agent resources to meet the user's search demand.

[0153] Condition 2: The search intent of the first search text is a valid intent.

[0154] After receiving the first search text sent by the client, the server may also use an intent analysis model or algorithm to perform intent analysis on the first search text to determine whether the search intent of the first search text is a valid intent.

[0155] For example, if the first search text is "How is the weather today?", the server can determine that the first search text has a clear scenario and can clearly express the user's intention without ambiguity. Then the server can perform the subsequent step S802, recall the agent and send it to the client, and the client displays the agent-related information to the user to meet the user's search requirements.

[0156] As for the first search text "Help me search", the server can determine that the first search text is incomplete and cannot constitute a clear and effective intention; as for the first search text "10, env", the server can determine that the first search text is meaningless text and cannot constitute a valid intention.

[0157] For these first search texts that cannot constitute clear and effective intentions, it is difficult for the server to recall intelligent agents in specific fields based on the first search texts to respond to user demands.

[0158] When the first search text satisfies condition 2, that is, when the search intent of the first search text is a valid intent, the field corresponding to the search intent of the first search text can be clarified, and then the server can execute subsequent step S802 to recall intelligent agents in a specific field to meet the user's search needs.

[0159] Condition 3: There is no search text matching the first search text in the search text cache data.

[0160] Each time the server obtains the question-and-answer results generated by the agent using the search text as a question, the server can cache the question-and-answer results corresponding to the search text.

[0161] When the server obtains the first search text, it can match the first search text with the historical search text in the text cache data. If there is a search text that matches the first search text, the server can directly recall the question and answer results and corresponding intelligent agent information corresponding to the search text that matches the first search text, without executing the subsequent process of matching the first search text with the existing intelligent agent.

[0162] If there is no search text matching the first search text in the text cache data, the server executes subsequent step S802, recalls the agent and sends it to the client, and the client displays the agent-related information to the user to meet the user's search request.

[0163] When the first search text satisfies condition 3, that is, when there is no search text matching the first search text in the search text cache data, it means that the server has not cached search texts or question-and-answer results with similar intent. The server can perform subsequent step S802 to recall the intelligent agent in a specific field to meet the user's search requirements.

[0164] In one possible implementation, after feeding back the information of each first agent and the question and answer results generated by each first agent to the client, the server may cache the search results in the following manner: update the search frequency corresponding to the search intent of the first search text; if the search frequency is greater than a preset frequency threshold, use the information of each first agent and the question and answer results generated by each first agent as the search results of the search intent, and cache the search results.

[0165] The server can record the corresponding search frequency according to the search intent of the search text, and when the search frequency is greater than the preset frequency threshold, cache the first agent corresponding to the current first search text and the question and answer results generated by each first agent as the search result of the search intent.

[0166] In this way, for the search intent that users frequently search, the server can directly recall the search results from the cache without repeatedly using the first search text to match the agent, nor does it need to repeatedly use the agent to generate question and answer results for different search texts with the same search intent. On the one hand, there is no need to generate question and answer results again, which shortens the waiting time for responding to user searches and improves the efficiency of responding to user searches, thereby improving user satisfaction; on the other hand, there is no need to repeatedly generate question and answer results for different search texts with the same search intent, which can save computing resources; on the other hand, feeding back the same question and answer results to users can ensure the consistency and quality of the agent's answers.

[0167] In a possible implementation, the server can also optimize the agent according to the agent's recall status and user interaction status. The present disclosure also discloses an agent optimization method, comprising the following steps: recording the recall information and user interaction information of each existing agent; optimizing the existing agents with optimization authority based on the recorded recall information and user interaction information; and / or generating optimization recommendation information based on the recorded recall information and user interaction information, and pushing the optimization recommendation information to the owner of the existing agent.

[0168] Recall information may include: the frequency of agents being recalled per unit time, the ranking of the matching degree between agents and search texts when they are recalled, etc. User interaction information may include: the frequency of users clicking on agents per unit time, the average duration of user-agent interaction, etc.

[0169] Based on this information, the server can optimize the models of existing agents with optimization permissions. For agents that do not have optimization permissions, the server can also push optimization recommendation information to the owners of these agents.

[0170] In this way, since the recall information and user interaction information reflect the user's satisfaction, optimizing the intelligent agent based on the above recall information and user interaction information and improving the ecological effect of the existing intelligent agent can also improve the user's satisfaction with the question-and-answer results generated by the intelligent agent to a certain extent. Furthermore, the higher the user satisfaction, the higher the user's willingness to interact with the intelligent agent, and the server can also obtain more user interaction information, thereby further optimizing the intelligent agent and achieving a positive feedback cycle.

[0171] In one possible implementation, for a question and answer result request sent by the client to the fourth agent, the server may provide feedback in the following manner: in response to the question and answer result request sent by the client, obtain the question and answer result generated by the agent indicated in the question and answer result request with the first search text as the question; and feed back the obtained question and answer result to the client.

[0172] The question-and-answer result request sent by the client may include the identification information of the agent. The server may obtain the agent identification in the question-and-answer result request, determine the agent indicated by the question-and-answer result request, and then obtain the question-and-answer result generated by the determined agent using the first search text as the question. Specifically, the server may input the first search text into the determined agent to obtain the question-and-answer result. After obtaining the question-and-answer result, the server may send the obtained question-and-answer result to the client.

[0173] In this way, when the user is not satisfied with the information of the first agent and the question and answer results generated by the first agent, the server can respond to the client's request for question and answer results, and feedback to the client the question and answer results generated by the fourth agent with the first search text as the question, so that the client can display the obtained question and answer results generated by the fourth agent to the user to meet the user's demand for further information.

[0174] Corresponding to the above-mentioned information search method applied to the client, the embodiment of the present disclosure provides an information search device applied to the client, such as Fig.11 As shown, the device comprises:

[0175] The sending module 1101 is used for sending the first search text to the server in response to the user inputting the first search text;

[0176] The first obtaining module 1102 is used to obtain information of multiple first agents matching the first search text and question-answering results generated by each first agent using the first search text as a question fed back by the server;

[0177] The display module 1103 is used to display the search results in the display interface using the information of each first intelligent agent and the question and answer results generated by each first intelligent agent as search results.

[0178] The solution provided by the embodiment of the present disclosure displays information of multiple agents and question-and-answer results generated by each first agent using the first search text as a question to the user according to the first search text input by the user, and can intuitively display the question-and-answer results to the user. Moreover, the question-and-answer results are generated by agents in a specific field, which attracts the user to further communicate with the agents, thereby satisfying the user's further search demands in a specific field.

[0179] In a possible implementation manner, the display module 1103 may be specifically used for:

[0180] The information of the second agent and the question and answer results generated by the second agent are displayed in the display interface, and the information of the third agent and the question and answer results generated by the third agent are hidden in the display interface, wherein the second agent includes: the first number of agents among multiple first agents in order of the matching degree with the first search text from high to low, and the third agent includes: the agents among the multiple first agents except the second agent.

[0181] In a possible implementation manner, the display module 1103 may also be used to:

[0182] In response to the hidden information expansion operation, the information of the third agent and the question and answer results generated by the third agent are displayed in the display interface.

[0183] In a possible implementation manner, the first obtaining module 1102 may also be used to:

[0184] Obtaining information of a plurality of fourth agents matched with the first search text fed back by the server, wherein the matching degree between the fourth agent and the first search text is less than the matching degree between the first agent and the first search text;

[0185] The display module 1103 can also be used to:

[0186] The information of the first agent and the question and answer results generated by the first agent are displayed in the display interface, and the information of the fourth agent is hidden in the display interface.

[0187] In a possible implementation manner, the information search device further includes:

[0188] A request module is used to respond to the hidden information launch operation and request the question and answer results for the fourth agent from the server; receive the question and answer results generated by the fourth agent using the first search text as the question from the server; and display the information of the fourth agent and the question and answer results generated by the fourth agent in the display interface.

[0189] Corresponding to the above-mentioned information search method applied to the server, the embodiment of the present disclosure provides an information search device applied to the server, such as Fig.12As shown, the device comprises:

[0190] A first acquisition module 1201 is used to acquire a first search text sent by a client;

[0191] The second obtaining module 1202 is used to obtain the matching degree between the first search text and each existing agent;

[0192] A recall module 1203 is used to recall the first second number of agents in the order of matching degree from high to low as the first agent;

[0193] The second acquisition module 1204 is used to obtain the question and answer results generated by each first agent using the first search text as the question;

[0194] The feedback module 1205 is used to feed back the information of each first agent and the question and answer results generated by each first agent to the client.

[0195] The solution provided by the embodiment of the present disclosure displays information of multiple agents and question-and-answer results generated by each first agent using the first search text as a question to the user according to the first search text input by the user, and can intuitively display the question-and-answer results to the user. Moreover, the question-and-answer results are generated by agents in a specific field, which attracts the user to further communicate with the agents, thereby satisfying the user's further search demands in a specific field.

[0196] In a possible implementation, the recall module 1203 can also be used to: recall the first third number of agents from the existing agents except the first agent in order of matching degree from high to low, as the fourth agent; and feed back information of each fourth agent to the client.

[0197] In a possible implementation manner, the second obtaining module 1202 may be specifically configured to:

[0198] For each existing intelligent agent, a first matching degree and / or a second matching degree is obtained between the first search text and the existing intelligent agent, and the matching degree between the first search text and the existing intelligent agent is determined based on the obtained matching degrees, wherein the first matching degree is: the matching degree between the first search text and the description information of the existing intelligent agent, and the second matching degree is: the matching degree between the field to which the first search text belongs and the field to which the existing intelligent agent belongs.

[0199] In a possible implementation manner, the second obtaining module 1202 may be specifically configured to:

[0200] If the obtained matching degree includes a first matching degree and a second matching degree, a weighted sum calculation is performed on the first matching degree and the second matching degree based on the weight coefficients corresponding to the first matching degree and the second matching degree respectively, and the weighted sum calculation result is used as the matching degree between the first search text and the existing intelligent agent.

[0201] In a possible implementation, the first matching degree may be obtained in the following manner:

[0202] Matching the keywords in the first search text with various types of description information of the existing agents to obtain the sub-matching degree corresponding to each type of description information;

[0203] Based on the weight coefficients corresponding to the respective types of description information, a weighted sum calculation is performed on the sub-matching degrees corresponding to the respective types of description information, and the weighted sum calculation result is used as the first matching degree.

[0204] In a possible implementation, each type of description information includes at least one of the following information: the name, introduction, recommendation words, and agent knowledge base of an existing agent.

[0205] In a possible implementation manner, the information search device may further include a judgment module, which is used to:

[0206] After obtaining the first search text sent by the client, it is determined whether the first search text satisfies at least one of the following conditions:

[0207] The first search text indicates searching for resources provided by a non-subject function provided by the client;

[0208] The search intent of the first search text is a valid intent;

[0209] There is no search text matching the first search text in the search text cache data.

[0210] In a possible implementation manner, the information search device may further include an optimization module, which is used to:

[0211] Record the recall information and user interaction information of each existing agent;

[0212] Based on the recorded recall information and user interaction information, the existing intelligent agents with optimization authority are optimized; and / or, based on the recorded recall information and user interaction information, optimization recommendation information is generated and pushed to the owners of the existing intelligent agents.

[0213] In a possible implementation manner, the information search device may further include a cache module, which is used to:

[0214] After feeding back the information of each first agent and the question-and-answer results generated by each first agent to the client, updating the search frequency corresponding to the search intent of the first search text;

[0215] If the search frequency is greater than a preset frequency threshold, the information of each first agent and the question and answer results generated by each first agent are used as search results of the search intent, and the search results are cached.

[0216] In a possible implementation manner, the feedback module 1205 may also be used to:

[0217] In response to a question-and-answer result request sent by a client, obtaining a question-and-answer result generated by the agent indicated in the question-and-answer result request using the first search text as the question;

[0218] Feedback the obtained question and answer results to the client.

[0219] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0220] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0221] Fig.13 A schematic block diagram of an example electronic device 1300 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein. The electronic device here can be an electronic device deployed with the above-mentioned client or server.

[0222] like Fig.13 As shown, the device 1300 includes a computing unit 1301, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1302 or a computer program loaded from a storage unit 1308 into a random access memory (RAM) 1303. In the RAM 1303, various programs and data required for the operation of the device 1300 can also be stored. The computing unit 1301, the ROM 1302, and the RAM 1303 are connected to each other via a bus 1304. An input / output (I / O) interface 1305 is also connected to the bus 1304.

[0223] A number of components in the device 1300 are connected to the I / O interface 1305, including: an input unit 1306, such as a keyboard, a mouse, etc.; an output unit 1307, such as various types of displays, speakers, etc.; a storage unit 1308, such as a disk, an optical disk, etc.; and a communication unit 1309, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1309 allows the device 1300 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0224] The computing unit 1301 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1301 performs the various methods and processes described above, such as the above-mentioned information search method. For example, in some embodiments, the above-mentioned information search method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1308. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1300 via the ROM 1302 and / or the communication unit 1309. When the computer program is loaded into the RAM 1303 and executed by the computing unit 1301, one or more steps of the above-mentioned information search method described above may be performed. Alternatively, in other embodiments, the computing unit 1301 may be configured to execute the above-mentioned information search method in any other appropriate manner (for example, by means of firmware).

[0225] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0226] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0227] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 foregoing.

[0228] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0229] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0230] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0231] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0232] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. An information search method, applied to a client, comprising: In response to a user inputting a first search text, sending the first search text to a server; Obtaining information of a plurality of first agents matched with the first search text and question-answer results generated by each first agent using the first search text as a question fed back by the server; The information of each first intelligent agent and the question and answer results generated by each first intelligent agent are used as search results, and the search results are displayed in a display interface.

2. The method according to claim 1, wherein: The step of taking the information of each first agent and the question-and-answer results generated by each first agent as search results and displaying the search results in a display interface includes: The information of the second agent and the question and answer results generated by the second agent are displayed in the display interface, and the information of the third agent and the question and answer results generated by the third agent are hidden in the display interface, wherein the second agent includes: the first number of agents among multiple first agents in order of the matching degree with the first search text from high to low, and the third agent includes: the agents among the multiple first agents except the second agent.

3. The method according to claim 2, further comprising: In response to the hidden information expansion operation, the information of the third agent and the question and answer results generated by the third agent are displayed in the display interface.

4. The method according to claim 1, further comprising: Obtaining information of a plurality of fourth agents matched with the first search text fed back by the server, wherein the matching degree between the fourth agents and the first search text is less than the matching degree between the first agents and the first search text; The step of taking the information of each first agent and the question-and-answer results generated by each first agent as search results and displaying the search results in a display interface includes: The information of the first agent and the question and answer results generated by the first agent are displayed in the display interface, and the information of the fourth agent is hidden in the display interface.

5. The method according to claim 4, further comprising: In response to the hidden information expansion operation, requesting the server for a question and answer result for the fourth agent; Receive the question and answer result generated by the fourth agent based on the first search text as the question fed back by the server; The information of the fourth agent and the question and answer results generated by the fourth agent are displayed in the display interface.

6. An information search method, applied to a server, comprising: Get the first search text sent by the client; Obtaining the matching degree between the first search text and each existing agent; In the order of matching degree from high to low, recall the second number of agents as the first agent; Obtaining question-and-answer results generated by each first agent using the first search text as a question; Feedback the information of each first agent and the question and answer results generated by each first agent to the client.

7. The method according to claim 6, further comprising: In the order of matching degree from high to low, recall the first third number of agents from the existing agents except the first agent as the fourth agent; Feedback the information of each fourth agent to the client.

8. The method according to claim 6 or 7, wherein: The obtaining of the matching degree between the first search text and each existing agent includes: For each existing agent, obtain a first matching degree and / or a second matching degree between the first search text and the existing agent, and determine the matching degree between the first search text and the existing agent based on the obtained matching degrees, wherein the first matching degree is: the matching degree between the first search text and the description information of the existing agent, and the second matching degree is: the matching degree between the field to which the first search text belongs and the field to which the existing agent belongs.

9. The method according to claim 8, wherein: The determining the matching degree between the first search text and the existing agent based on the obtained matching degree includes: If the obtained matching degree includes the first matching degree and the second matching degree, a weighted sum calculation is performed on the first matching degree and the second matching degree based on the weight coefficients corresponding to the first matching degree and the second matching degree respectively, and the weighted sum calculation result is used as the matching degree between the first search text and the existing intelligent agent.

10. The method according to claim 8, wherein: The first matching degree is obtained in the following manner: Matching the keywords in the first search text with various types of description information of the existing agents to obtain sub-matching degrees corresponding to the various types of description information; Based on the weight coefficients corresponding to the respective types of description information, a weighted sum calculation is performed on the sub-matching degrees corresponding to the respective types of description information, and the weighted sum calculation result is used as the first matching degree.

11. The method according to claim 10, wherein: Each type of description information includes at least one of the following information: the name, introduction, recommendation words, and knowledge base of the existing intelligent agent.

12. The method according to claim 6 or 7, after obtaining the first search text sent by the client, further comprising: Determine whether the first search text satisfies at least one of the following conditions. If so, execute the step of obtaining the matching degree between the first search text and each existing agent: The first search text indicates searching for resources provided by the non-subject function provided by the client; The search intent of the first search text is a valid intent; There is no search text matching the first search text in the search text cache data.

13. The method according to claim 6 or 7, further comprising: Record the recall information and user interaction information of each existing agent; Based on the recorded recall information and user interaction information, optimize the existing agents with optimization authority; And / or, based on the recorded recall information and user interaction information, generate optimization recommendation information, and push the optimization recommendation information to the owner of the existing intelligent agent.

14. The method according to claim 6 or 7, after feeding back the information of each first agent and the question-answering results generated by each first agent to the client, further comprising: Updating the search frequency corresponding to the search intent of the first search text; If the search frequency is greater than a preset frequency threshold, the information of each first agent and the question and answer results generated by each first agent are used as search results of the search intention, and the search results are cached.

15. The method according to claim 6 or 7, further comprising: In response to the question and answer result request sent by the client, obtaining the question and answer result generated by the agent indicated by the question and answer result request using the first search text as the question; The obtained question and answer result is fed back to the client.

16. An information search device, applied to a client, comprising: A sending module, configured to send a first search text to a server in response to a user inputting the first search text; A first acquisition module, used to obtain information of multiple first agents matched with the first search text and question-answer results generated by each first agent using the first search text as a question fed back by the server; A display module is used to use the information of each first intelligent agent and the question and answer results generated by each first intelligent agent as search results, and display the search results in a display interface.

17. The device according to claim 16, wherein: The display module is specifically used for: The information of the second agent and the question and answer results generated by the second agent are displayed in the display interface, and the information of the third agent and the question and answer results generated by the third agent are hidden in the display interface, wherein the second agent includes: the first number of agents among multiple first agents in order of the matching degree with the first search text from high to low, and the third agent includes: the agents among the multiple first agents except the second agent.

18. The device according to claim 17, wherein: The display module is also used for: In response to the hidden information expansion operation, the information of the third agent and the question and answer results generated by the third agent are displayed in the display interface.

19. The device according to claim 16, wherein: The first obtaining module is further used for: Obtaining the server-side feedback that matches the first search text: Obtaining information of a plurality of fourth agents matched with the first search text fed back by the server, wherein the matching degree between the fourth agents and the first search text is less than the matching degree between the first agents and the first search text; The display module is also used for: The information of the first agent and the question and answer results generated by the first agent are displayed in the display interface, and the information of the fourth agent is hidden in the display interface.

20. The apparatus according to claim 19, further comprising: A request module, configured to request a question and answer result from the server for the fourth agent in response to the hidden information expansion operation; Receive the question and answer results generated by the fourth agent using the first search text as the question fed back by the server; and display the information of the fourth agent and the question and answer results generated by the fourth agent in the display interface.

21. An information search device, applied to a server, comprising: A first acquisition module, used to acquire a first search text sent by a client; A second obtaining module, used to obtain the matching degree between the first search text and each existing intelligent agent; A recall module, used for recalling the second number of agents as the first agent in the order of matching degree from high to low; A second acquisition module is used to obtain the question and answer results generated by each first agent using the first search text as the question; A feedback module is used to feed back information of each first agent and question-answering results generated by each first agent to the client.

22. The device according to claim 21, wherein The recall module is also used for: In the order of matching degree from high to low, recall the first third number of agents from the existing agents except the first agent as the fourth agent; Feedback the information of each fourth agent to the client.

23. The device according to claim 21 or 22, wherein: The second obtaining module is specifically used for: For each existing agent, obtain a first matching degree and / or a second matching degree between the first search text and the existing agent, and determine the matching degree between the first search text and the existing agent based on the obtained matching degrees, wherein the first matching degree is: the matching degree between the first search text and the description information of the existing agent, and the second matching degree is: the matching degree between the field to which the first search text belongs and the field to which the existing agent belongs.

24. The device according to claim 23, wherein: The second obtaining module is specifically used for: If the obtained matching degree includes the first matching degree and the second matching degree, a weighted sum calculation is performed on the first matching degree and the second matching degree based on the weight coefficients corresponding to the first matching degree and the second matching degree respectively, and the weighted sum calculation result is used as the matching degree between the first search text and the existing intelligent agent.

25. The device according to claim 23, wherein: The first matching degree is obtained in the following manner: Matching the keywords in the first search text with various types of description information of the existing agents to obtain sub-matching degrees corresponding to the various types of description information; Based on the weight coefficients corresponding to the respective types of description information, a weighted sum calculation is performed on the sub-matching degrees corresponding to the respective types of description information, and the weighted sum calculation result is used as the first matching degree.

26. The device according to claim 25, wherein Each type of description information includes at least one of the following information: the name, introduction, recommendation words, and knowledge base of the existing intelligent agent.

27. The device according to claim 21 or 22, further comprising a judgment module, configured to: after obtaining the first search text sent by the client, judge whether the first search text satisfies at least one of the following conditions: The first search text indicates searching for resources provided by the non-subject function provided by the client; The search intent of the first search text is a valid intent; There is no search text matching the first search text in the search text cache data.

28. The device according to claim 21 or 22, wherein: The device also includes an optimization module, which is used to: Record the recall information and user interaction information of each existing agent; Based on the recorded recall information and user interaction information, optimize the existing agents with optimization authority; And / or, based on the recorded recall information and user interaction information, generate optimization recommendation information, and push the optimization recommendation information to the owner of the existing intelligent agent.

29. The device according to claim 21 or 22, further comprising a cache module, configured to: After feeding back the information of each first agent and the question-and-answer results generated by each first agent to the client, updating the search frequency corresponding to the search intent of the first search text; If the search frequency is greater than a preset frequency threshold, the information of each first agent and the question and answer results generated by each first agent are used as search results of the search intention, and the search results are cached.

30. The device according to claim 21 or 22, wherein: The feedback module is further used for: In response to the question and answer result request sent by the client, obtaining the question and answer result generated by the agent indicated by the question and answer result request using the first search text as the question; The obtained question and answer result is fed back to the client.

31. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5 or 6-15.

32. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-5 or 6-15.

33. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1-5 or 6-15.

Images