Information search method and apparatus

Abstract

The one or more embodiments of the specification provide an information search method and device. The method comprises: obtaining input data from a target object through a unified search entrance, the unified search entrance corresponding to vertical search engines of multiple dimensions; performing semantic analysis on the input data to determine a target dimension matching a search intention of the input data; and calling a vertical search engine of the target dimension to perform a search operation.

Description

Technical Field

[0001] This specification relates to one or more embodiments in the field of information processing technology, and in particular to an information search method and apparatus. Background Technology

[0002] Mobile office software has gradually become an important tool for enterprise organization management, communication, collaboration, and work coordination. With the increasing functionality of mobile office software, data search plays a crucial role in daily office work and is also an important tool for improving work efficiency.

[0003] In related technologies, different search contents correspond to different search interfaces. When conducting cross-dimensional content searches, users need to switch between multiple search pages and enter the corresponding search terms to search. Summary of the Invention

[0004] In view of the above, one or more embodiments of this specification provide an information search method and apparatus.

[0005] To achieve the above objectives, one or more embodiments of this specification provide the following technical solutions:

[0006] According to a first aspect of one or more embodiments of this specification, an information search method is proposed, comprising:

[0007] Input data from the target object is obtained through a unified search portal, which corresponds to a vertical search engine with multiple dimensions;

[0008] Perform semantic analysis on the input data to determine the target dimension that matches the search intent of the input data;

[0009] The vertical search engine for the target dimension is invoked to perform a search operation.

[0010] According to a second aspect of one or more embodiments of this specification, an information search device is provided, comprising:

[0011] The acquisition module is used to acquire input data from the target object through a unified search entry point, which corresponds to a vertical search engine with multiple dimensions.

[0012] The determination module is used to perform semantic analysis on the input data to determine the target dimension that matches the search intent of the input data;

[0013] The calling module is used to invoke the vertical search engine for the target dimension to perform search operations.

[0014] According to a third aspect of one or more embodiments of this specification, an electronic device is provided, comprising:

[0015] processor;

[0016] Memory used to store processor-executable instructions;

[0017] The processor implements the method described above by running the executable instructions.

[0018] According to a fourth aspect of one or more embodiments of this specification, a computer-readable storage medium is provided that stores computer instructions thereon, which, when executed by a processor, implement the steps of the method described in any of the preceding claims.

[0019] The technical solutions provided in the embodiments of this specification may include the following beneficial effects:

[0020] In this embodiment of the specification, before performing a search operation, the search engine performs semantic analysis on the input data to determine the search intent, rather than directly using the input data as search terms. This eliminates the need for users to accurately input keywords corresponding to their search intent. Even if users input complex, long sentences, accurate search results can be obtained through the identification of the search intent. Furthermore, a unified search entry point corresponding to multiple vertical search engines is provided. For search needs across multiple dimensions, users can input data on a unified search page (e.g., the application's homepage), achieving cross-dimensional searches without switching between search pages. This simplifies data input and improves the search experience.

[0021] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the architecture of an information search system provided in an exemplary embodiment.

[0023] Figure 2a This is a flowchart of an exemplary embodiment of an information search method.

[0024] Figure 2b This is a schematic diagram illustrating the correspondence between a unified search entry point, input components, and a vertical search engine, provided in an exemplary embodiment.

[0025] Figure 2c This is a schematic diagram of a user interface provided in an exemplary embodiment.

[0026] Figure 2d This is a schematic diagram of another user interface provided in an exemplary embodiment.

[0027] Figure 2eThis is a schematic diagram of yet another user interface provided in an exemplary embodiment.

[0028] Figure 3 This is a flowchart of another information search method provided in an exemplary embodiment.

[0029] Figure 4 This is a schematic diagram of a session system used in an information search method, provided as an exemplary embodiment.

[0030] Figure 5 This is a schematic diagram of a dialog-based interactive page provided in an exemplary embodiment.

[0031] Figure 6 This is an exemplary embodiment of a page that provides feedback reminder information to a target object.

[0032] Figure 7 This is a schematic structural diagram of a device provided in an exemplary embodiment.

[0033] Figure 8 This is a block diagram of an information search device provided in an exemplary embodiment. Detailed Implementation

[0034] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0035] It should be noted that the steps of the corresponding methods are not necessarily performed in the order shown and described in this specification in other embodiments. In some other embodiments, the methods may include more or fewer steps than described in this specification. Furthermore, a single step described in this specification may be broken down into multiple steps in other embodiments; and multiple steps described in this specification may be combined into a single step in other embodiments.

[0036] In one embodiment, the information search scheme of this specification can be applied to electronic devices, such as mobile phones, tablets, laptops, PDAs (Personal Digital Assistants), wearable devices (such as smart glasses, smartwatches, etc.), and this specification does not limit this. During operation, the electronic device can obtain user input data through human-computer interaction and perform information search based on the input data.

[0037] Figure 1 This is a schematic diagram of the architecture of an information search system provided in an exemplary embodiment. For example... Figure 1 As shown, the system may include a server 11, a network 12, and several electronic devices, such as mobile phones 13, 14, and 15.

[0038] Server 11 can be a physical server containing an independent host, or it can be a virtual server hosted in a host cluster. During operation, server 11 can run server-side programs for a specific application to implement its related functions. For example, when server 11 runs a mobile group office platform program, it can act as the server for that platform. In one or more embodiments of this specification, server 11 can cooperate with clients running on mobile phones 13-15 to implement an information search system solution.

[0039] In this embodiment, the mobile group office platform can not only realize communication functions, but also serve as an integrated functional platform for many other functions, such as processing internal group events (e.g., leave requests, office supplies requisitions, financial approvals), attendance events, task events, log events, etc., and processing external group events such as ordering meals and purchasing. One or more embodiments of this specification do not limit this.

[0040] Mobile phones 13-15 are just one type of electronic device that users can use. In reality, users can obviously also use electronic devices such as tablets, laptops, PDAs (Personal Digital Assistants), wearable devices (such as smart glasses, smartwatches, etc.), etc., and one or more embodiments in this specification do not limit this. During operation, the electronic device can run a client-side program of an application to implement the relevant application functions. For example, when the electronic device runs a program for a mobile group office platform, it can act as a client of that mobile group office platform.

[0041] It should be noted that the client application of the mobile group office platform can be pre-installed on electronic devices, allowing the client to be launched and run on those devices. Of course, when using online "clients" employing technologies such as HTML5 (HyperText Markup Language 5), it is not necessary to install the corresponding application on the electronic device to obtain and run the client. As for the network 12 that facilitates interaction between mobile phones 13-15 and server 11, it can include various types of wired or wireless networks. In one embodiment, network 12 can include the Public Switched Telephone Network (PSTN) and the Internet.

[0042] Figure 2a This is a flowchart of an exemplary embodiment of an information search method applied to a terminal. The method may include the following steps:

[0043] Step 202: Obtain input data from the target object through a unified search portal.

[0044] The target audience could be, for example, users who have information search needs.

[0045] In this embodiment, input data is obtained through a unified search entry point, see [link to relevant documentation]. Figure 2b The unified search portal can interface with input boxes, predefined options, microphones, cameras, and other input components in the user interface displayed on the terminal, thereby providing multiple data input channels for the target object. The unified search portal can be implemented, but is not limited to, through an Application Programming Interface (API).

[0046] It should be noted that components such as cameras and microphones do not necessarily have to be installed on the terminal. They can also utilize the capabilities of other devices that can establish a communication connection with the terminal, such as smart speakers, television cameras / microphones, in-vehicle cameras / microphones in a car environment, wearable devices, etc. For example, if a user launches the mobile group office platform client and enters data through the client's user interface, Figure 2cThis is a schematic diagram of a user interface provided in an exemplary embodiment. The target object can input text data into input boxes on the user interface, and the unified search entry can receive this text data as input data for information search. The target object can trigger a "press and hold to speak" action to activate the terminal microphone and input audio data through the microphone. Alternatively, the target object can input audio data through the microphone of another device, which will then send the audio data to the terminal. The unified search entry can collect this audio data as input data for information search. The target object can trigger predefined options displayed on the user interface. For example, when the "Meeting Records Query" option is triggered, the unified search entry can use the predefined data corresponding to the triggered "Meeting Records Query" option as input data for information search. These predefined options can be, but are not limited to, displayed in the user interface as button controls. Each button control can display text content generated by a recommendation algorithm, and the text content corresponds to the predefined data. The target object can also make specific facial expressions, gestures, or actions, and the unified search entry can capture images or videos of the target object through the terminal's camera or other device's camera, using these images or videos as input data. Therefore, users can input data containing the content to be searched through various input methods such as voice input, button control triggering, spelling input, and making specific expressions, gestures, and actions.

[0047] See Figure 2b The unified search portal also corresponds to multiple vertical search engines. These vertical search engines can provide users with search services for different application types and fields. For example, they can search logs, weekly reports, set tasks, contact information, historical chat records, call records, approval records, and so on. It's important to note that one vertical search engine can correspond to one dimension, meaning each vertical search engine performs content search within that dimension; or multiple vertical search engines can correspond to a single dimension, such as configuring two vertical search engines for log searches and three vertical search engines for approval record searches, etc.

[0048] Step 204: Perform semantic analysis on the input data to determine the target dimension that matches the search intent of the input data.

[0049] Semantic analysis is generally based on text-based input data. Therefore, if the input data contains non-text data, it must be converted into text data before semantic analysis can be performed. For example, if the input data contains audio data, it can be converted into text data using, but is not limited to, Automatic Speech Recognition (ASR) technology.

[0050] Performing semantic analysis on the input data is to determine the search intent of the target object, that is, to determine the application type and field of the search service required by the target object, and then a vertical search engine for the target dimension matching the search intent can be selected to perform the search operation.

[0051] Each vertical search engine is pre-configured with semantic slots matching its search dimension. The semantic slots contain at least one slot, and the determination of the target dimension can be achieved with the help of semantic slots. Determining the target dimension matching the search intent may include the following steps:

[0052] S1. Perform word segmentation on the input data.

[0053] In another embodiment, after word segmentation, stop words in the word segmentation result can also be removed. Among them, stop words can be, for example, meaningless words such as "ah", "ne", "de", punctuation marks, special symbols, etc. For example, if the input data is "Find the meeting record of Zhou Moumou", after performing word segmentation on it, the word segmentation result is "Find", "Zhou Moumou", "de", "meeting record". Among them, "de" is meaningless and can be removed from the word segmentation result. The final word segmentation result is "Find", "Zhou Moumou", "meeting record".

[0054] S2. Fill each phrase in the word segmentation result into each slot matching the slot type of the semantic slot.

[0055] Suppose there are two vertical search engines. Among them, the semantic slot of vertical search engine a contains two slots: the item (event) slot and the who (object) slot; the semantic slot of vertical search engine b contains two slots: the location (location) slot and the time (time) slot. Fill the phrases in the above word segmentation result into the semantic slots of vertical search engine a and vertical search engine b respectively. The filling results are shown in Table 1 and Table 2.

[0056] Table 1

[0057] slot Slot value item Meeting minutes who Zhou Moumou

[0058] Table 2

[0059] slot Slot value location time

[0060] S3. Determine the target dimension according to the filling result.

[0061] As can be seen from Table 1 and Table 2, the phrases in the word segmentation result can fill the slots of the semantic slot of vertical search engine a, but cannot fill the slots of the semantic slot of vertical search engine b. Thus, it can be judged that the search intent of the input data matches the dimension of vertical search engine a and does not match vertical search engine b.

[0062] The slot filling rules can be set according to actual needs. For example, it can be set to confirm that the search intent of the input data matches the dimension of the vertical search engine corresponding to the semantic slot only when the word segmentation results can fill all the slots of the semantic slot. Alternatively, it can be set to confirm that the search intent of the input data matches the dimension of the vertical search engine corresponding to the semantic slot when the word segmentation results can fill a preset number of slots in the semantic slot.

[0063] Understandably, if the input data can satisfy the slot filling rules of multiple semantic slots, it means that the search intent of the input data matches the vertical search engines of multiple dimensions. In this case, the vertical search engines of multiple dimensions can be selected as the vertical search engine of the target dimension for search operation.

[0064] Step 206: Call the vertical search engine of the target dimension to perform a search operation.

[0065] When a vertical search engine for a target dimension performs a search operation, it determines the phrases that fill the slots in the semantic slots as search terms and performs a search operation based on those search terms.

[0066] Search results obtained by a vertical search engine can be displayed in, but are not limited to, the following formats: text, audio, and predefined options, where each predefined option corresponds to a search result. For audio, text-based search results can be converted into corresponding acoustic signals using text-to-speech (TTS).

[0067] In this embodiment, before performing a search operation, the search engine performs semantic analysis on the input data to determine the search intent, rather than directly using the input data as search terms. This eliminates the need for users to accurately input keywords corresponding to their search intent. Even if users input complex, long sentences, accurate search results can be obtained through the identification of the search intent. Furthermore, a unified search entry point corresponding to multiple vertical search engines is provided. For search needs across different application types, users can input data on a unified search page (e.g., the application's homepage), achieving cross-application type searches without switching between search pages. This simplifies data input and improves the search experience.

[0068] In another embodiment, if there are multiple vertical search engines for target dimensions that match the search intent of the input data, the multiple vertical search engines for target dimensions will obtain multiple search results when performing search operations. For example, suppose the input data is "find Zhou Moumou", and the search intent of the input data may be to find Zhou Moumou's logs, contact information, or leave approval requests. If there are 3 vertical search engines for different dimensions that match it, calling these 3 vertical search engines may result in 3 search results.

[0069] A vertical search engine in a single dimension may also yield multiple search results. For example, suppose the input data is "find Zhou Moumou's approval records". We can determine that the search intent of the input data is to find approval records. The input data identifies the object to which the approval records belong, but it does not determine when the approval records to be found were generated. Therefore, the vertical search engine may obtain multiple search results.

[0070] In cases where multiple search results are obtained, the user's historical behavior data can be analyzed to determine the user's search preferences and search history habits. The search results can be preprocessed before being displayed. Preprocessing can include, but is not limited to, filtering or sorting the search results.

[0071] Let's take the input data "Find Zhou Moumou" as an example. Assume there are three vertical search engines matching this input data, resulting in three search results. Analyzing the target's historical behavior data reveals that the target frequently views Zhou Moumou's contact information. Therefore, Zhou Moumou's contact information can be filtered from the three search results and displayed, while Zhou Moumou's logs and leave approval requests are not displayed. Alternatively, the three search results can be sorted, see [link to relevant documentation]. Figure 2d Search results will be displayed in the following order: contact information, meeting minutes, and leave approval requests.

[0072] For cases where multiple search results are obtained, these results can be displayed corresponding to vertical search engines targeting different dimensions. Taking the input data "find Zhou Moumou" as an example, if the search results include Zhou Moumou's phone number, instant messaging account, email address, meeting minutes from July 3rd, and meeting minutes from July 10th; the phone number, instant messaging account, and email address are obtained from the vertical search engine targeting contact information, while the meeting minutes from July 3rd and July 10th are obtained from the vertical search engine targeting meeting minutes. See [link to relevant documentation]. Figure 2e The system can display Zhou's phone number, instant messaging account, and email address in one display area, and the meeting minutes from July 3rd and July 10th in another. It should be noted that the display method of the search results is not limited. Figure 2eAlternatively, the phone number, instant messaging account, and email address can be displayed in a floating window, while the meeting minutes from July 3rd and July 10th can be displayed in another floating window.

[0073] Figure 3 This is a flowchart of another information search method provided in an exemplary embodiment. In this embodiment, information search is taken as an example on a mobile group office platform. See [link to flowchart]. Figure 3 The method may include the following steps:

[0074] Step 302: Obtain input data from the target object through a unified search portal.

[0075] The unified search entry point corresponds to multiple vertical search engines, each offering users search services for different application types and fields. The mobile team office platform can, but is not limited to, providing services such as log creation, weekly report creation, task setting, address book access, schedule creation, and approvals. Correspondingly, the different vertical search engines can be used to search for logs, weekly reports, set tasks, contact information, historical chat logs, call logs, schedules, and approval records, among other things.

[0076] In this embodiment, the unified search entry point can also interface with input boxes, predefined options, microphones, cameras, and other components in the user interface of the mobile group office platform client, providing multiple data input channels for the target object. Thus, the target object can input data containing the search query content through various methods, such as input boxes in the user interface of the mobile group office platform client, predefined options displayed in the user interface, and microphone activation controls.

[0077] For example, if a user launches the mobile group office platform client and enters data in the client's user interface through input operations, such as "Schedule a meeting with Mr. / Ms. Zhou at 3 PM" via voice input, the unified search entry will retrieve the input data "Schedule a meeting with Mr. / Ms. Zhou at 3 PM" and send the input data to the conversation system for semantic analysis.

[0078] Step 304: Call the session system to perform semantic analysis on the input data.

[0079] In this embodiment, the conversation system can perform semantic analysis on the data generated by the target object's recent input operations and the data generated by its historical input operations to determine the target dimension that matches the search intent of the input data. The data generated by recent input operations and the data generated by historical input operations refer to the data generated by the target object's input operations during the dialogue interaction.

[0080] In this embodiment, if the search intent is unclear, multiple rounds of dialogue interaction can be conducted with the target object, and the search intent can be determined based on the data generated by the target object's input operations during the multiple rounds of dialogue interaction, thereby determining the target dimension that matches the search intent.

[0081] The conversational system can be implemented, but is not limited to, through a smart assistant. Figure 4 This is a schematic diagram of a conversation system provided in an exemplary embodiment. The conversation system includes an NLU (Natural Language Understanding) module, a DM (Dialogue Management) module, and an NLG (Natural Language Generation) module.

[0082] The NLU module is used to analyze and process input data and session context through computer algorithms or models, transforming them into structured information and determining the intent of the user's input. The session context refers to the data generated by the target object's input operations during multi-turn dialogue interactions.

[0083] DM module: Used to select appropriate system actions (including asking, replying, and performing search operations) based on the intent identified by the NLU module and the session context.

[0084] The NLG module is used to convert the system actions output by the DM module into dialog prompts when the system action is a query or response; and to call the corresponding vertical search engine when the system action is to perform a search operation.

[0085] For example, if a user enters "I want to find Zhou Moumou's phone number", semantic analysis can be performed to clarify the search intent. In this case, there is no need to generate a dialog prompt.

[0086] If a user enters "I want to find Zhou Moumou", their search intent could be "to find contact information", "to search for chat history", or "to search for weekly reports", etc., making it difficult to specify the search intent. In this case, a dialog prompt can be generated to guide the target user to enter more input data that clearly defines the search intent. The input data obtained through multiple rounds of dialogue interaction can clarify the search intent.

[0087] During interactive dialogue, dialogue prompts may be displayed in at least one of the following forms, but not limited to: text, voice, and predefined options, where each predefined option corresponds to a dialogue prompt. For each round of dialogue, the user can input data using at least one of the following input methods: entering text data in an input box, inputting audio data via microphone, or triggering a predefined option.

[0088] For example Figure 5This is a schematic diagram of a dialog-based interactive page provided in an exemplary embodiment. See also: Figure 5 The user's initial input is "I want to find Zhou Moumou". Since the search intent cannot be determined based on this input, the conversation system can generate a dialog prompt. Figure 5 The dialogue prompts are displayed in the form of predefined options, each corresponding to a dialogue prompt content. When the predefined option "make a phone call" is triggered, the predefined data "contact information" corresponding to the predefined option is used as the input data of the target object in this round of dialogue. When performing intent recognition, the search intent is determined to be "find Zhou Moumou's contact information" based on the data generated by the target object's historical input operation "I want to find Zhou Moumou" and the data "contact information" generated by the most recent input operation. Figure 5 The demonstrated dialogue interaction clarifies the search intent through two rounds of dialogue. In actual use, if the search intent cannot be clarified through two rounds of dialogue, further dialogue prompts can be generated to guide the target object to input data, and the search intent can be clarified through data from multiple rounds of dialogue interaction.

[0089] Step 306: Determine whether to perform a search operation based on the semantic analysis results.

[0090] If the search intent of the input data can be determined through semantic analysis, then proceed to step 308.

[0091] If semantic analysis determines that the intent of the input data does not involve information search, such as the client's intention to close the app, then the operation of calling the vertical search engine will not be executed; instead, other operations corresponding to the intent will be executed.

[0092] Step 308: Determine the target dimension that matches the search intent, and call the vertical search engine of the target dimension to perform the search operation.

[0093] When a vertical search engine for a target dimension performs a search operation, it determines the phrases that fill the slots in the semantic slots as search terms and performs a search operation based on those search terms.

[0094] Search results obtained by a vertical search engine can be displayed in, but are not limited to, the following formats: text, audio, and predefined options, where each predefined option corresponds to a search result. For audio, text-based search results can be converted into corresponding acoustic signals using text-to-speech (TTS).

[0095] In another embodiment, if multiple search results are obtained, they can be preprocessed based on the target object's historical behavior data before being displayed. The user's historical behavior data is a record of the user's historical operations (e.g., viewing, clicking, etc.) on the mobile group office platform client. Multiple search results may be obtained for two reasons: there are multiple vertical search engines matching the search intent of the input data; or a single-dimensional vertical search engine may also yield multiple search results. For the aforementioned situation of obtaining multiple search results, the user's historical behavior data can be analyzed to determine the user's search preferences and search history habits, and the search results can be preprocessed before being displayed. For the specific implementation process, please refer to the description of the corresponding part of the information search method embodiment shown in Figure 2, which will not be repeated here.

[0096] In another embodiment, if the search results obtained by the vertical search engine for the target dimension conflict with the target operation, a reminder message can be sent to the target user. The input data is related to the target operation, which can be, but is not limited to, setting up schedules or creating tasks on the mobile group office platform client. For example, if a user inputs "Schedule a meeting with Zhou at 3 PM," semantic analysis determines that the user's intent is meeting scheduling (the target operation). The target dimension matching the search intent of the input data is "meeting scheduling query." The session system then calls the vertical search engine for the meeting scheduling query dimension to perform a search. If the search reveals that the user has already scheduled a meeting with other colleagues at the same time slot, it indicates a conflict between the user's target operation and the search results, and the session system sends a reminder message. Figure 6 This is an exemplary embodiment of a page that provides feedback reminder information to a target object. The reminder information may be, for example, […]. Figure 6 The system displays a reminder message: "You have already scheduled a meeting with Mr. Liu at 3 PM. Do you want to continue making the appointment?" At the same time, to facilitate the user's subsequent input, it can also provide dialog prompts in the form of predefined options such as "OK" and "Cancel".

[0097] Figure 7 This is a schematic structural diagram of a device provided in an exemplary embodiment. Please refer to... Figure 7At the hardware level, the device includes a processor 702, an internal bus 704, a network interface 706, memory 708, and non-volatile memory 710, and may also include other hardware required by the application. The processor 702 reads the corresponding computer program from the non-volatile memory 710 into the memory 708 and then runs it, forming an information search device at the logical level. Of course, in addition to the software implementation, one or more embodiments of this specification do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0098] Please refer to Figure 8 In a software implementation, the information search device may include:

[0099] The acquisition module 81 is used to acquire input data from the target object through a unified search entry, which corresponds to a vertical search engine with multiple dimensions.

[0100] The determination module 82 is used to perform semantic analysis on the input data and determine the target dimension that matches the search intent of the input data.

[0101] Module 83 is invoked to invoke the vertical search engine for the target dimension to perform a search operation.

[0102] Optionally, the input data is obtained through the unified search portal, including at least one of the following:

[0103] Receive the text data entered by the target object in the input box corresponding to the unified search entry;

[0104] Collect the audio data input by the target object to the unified search entry point;

[0105] Display predefined options corresponding to the unified search entry, and use the predefined data corresponding to the triggered predefined options as input data from the target object.

[0106] Optionally, the device further includes:

[0107] A conversion module is used to convert non-text data into text data when the input data contains non-text data, so as to perform semantic analysis on the text data.

[0108] Optionally, the input data includes:

[0109] The data generated by the most recent input operation of the target object; or, the data generated by the most recent input operation of the target object and the data generated by historical input operations.

[0110] Optionally, each vertical search engine is pre-configured with semantic slots that match its search dimensions, and the semantic slots contain at least one slot.

[0111] The determining module is specifically used for:

[0112] The input data is processed by word segmentation;

[0113] Each word group in the word segmentation result is filled into a slot that matches the slot type of the semantic slot.

[0114] The target dimension is determined based on the word-filling results.

[0115] Optionally, the calling module is specifically used for:

[0116] The phrases that fill the slots in the semantic slots are determined as search terms;

[0117] Invoke the vertical search engine for the target dimension to perform a search operation for the search term.

[0118] Optionally, the device further includes:

[0119] A generation module is used to generate dialogue prompts to guide the target object to input input data containing word groups that match the slot types of unfilled word groups.

[0120] Optionally, if search results are obtained from multiple vertical search engines targeting different dimensions, or if multiple search results are obtained from a single target search engine, the device further includes:

[0121] The acquisition module is used to acquire historical behavior data of the target object;

[0122] The display module is used to sort multiple search results based on the historical behavior data and display the sorted search results; or, to filter the multiple search results based on the historical behavior data and display the filtered search results.

[0123] Optionally, the input data is related to the target operation; the device further includes:

[0124] The feedback module is used to provide a reminder to the target object when the search results obtained by the vertical search engine in the target dimension conflict with the target operation.

[0125] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which can take the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email sending and receiving device, game console, tablet computer, wearable device, or any combination of these devices.

[0126] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0127] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0128] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can be implemented by any method or technology to store information, on which a computer program (information) is stored, which, when executed by a processor, implements the method steps provided in any of the above embodiments. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0129] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0130] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0131] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of one or more embodiments of this specification. The singular forms “a,” “described,” and “the” used in one or more embodiments of this specification and in the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0132] It should be understood that although the terms first, second, third, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of one or more embodiments of this specification, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "in response to a determination," or "when," or "in the event of a determination."

[0133] The above description is merely a preferred embodiment of one or more embodiments of this specification and is not intended to limit the scope of one or more embodiments of this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this specification should be included within the protection scope of one or more embodiments of this specification.

Claims

1. An information search method characterized by comprising: The method comprises the following steps: obtaining input data from a target object through a unified search portal, wherein the unified search portal corresponds to a plurality of vertical search engines of multiple dimensions; each vertical search engine is pre-configured with a semantic slot matching the search dimension thereof, and the semantic slot comprises at least one slot; performing word segmentation processing on the input data; filling each word group in the word segmentation result into each slot matching the slot type of the semantic slot; wherein the slot filling rule comprises: if the word segmentation result can fill all slots of the semantic slot, it is determined that the search intention of the input data matches the dimension of the vertical search engine corresponding to the semantic slot; or if the word segmentation result can fill a preset number of slots of the semantic slot, it is determined that the search intention of the input data matches the dimension of the vertical search engine corresponding to the semantic slot; determining the search dimension of the vertical search engine corresponding to the semantic slot which can be filled with all slots or a preset number of slots as the target dimension matching the search intention of the input data; determining the word group filled in the slot of the semantic slot as a search word, and calling the vertical search engine of the target dimension to perform a search operation on the search word.

2. The information search method according to claim 1, characterized by, Obtaining the input data through the unified search portal comprises at least one of the following: receiving text data input by the target object in an input box corresponding to the unified search portal; collecting audio data input by the target object to the unified search portal; displaying predefined options corresponding to the unified search portal, and inputting predefined data corresponding to the triggered predefined option as input data from the target object.

3. The information search method according to Claim 1, characterized by, Further comprising: in the case that the input data comprises non-text type data, converting the non-text type data into text type data for semantic analysis on the text type data.

4. The information search method according to Claim 1, characterized by, The input data comprises: data generated by a recent input operation of the target object, or data generated by a recent input operation of the target object and data generated by a historical input operation.

5. The information search method according to Claim 1, characterized by, In the case that the slot of the semantic slot is not filled, the method further comprises: generating a dialogue prompt to guide the target object to input input data containing a word group matching the slot type of the unfilled slot.

6. The information search method according to Claim 1, characterized by, In the case that the vertical search engines of multiple target dimensions all obtain search results or one target search engine obtains multiple search results, the method further comprises: obtaining historical behavior data of the target object; sorting the multiple search results according to the historical behavior data and displaying the sorted search results, or filtering the multiple search results according to the historical behavior data and displaying the filtered search results.

7. The information search method according to Claim 1, characterized by, In the case that the vertical search engines of multiple target dimensions all obtain search results, the method further comprises: corresponding display of the multiple search results and the vertical search engines of the multiple target dimensions.

8. The information search method according to Claim 1, characterized by, The input data is related to a target operation, and the method further comprises: In the case that the search result obtained by the vertical search engine of the target dimension conflicts with the target operation, feedback reminding information is provided to the target object.

9. An information search apparatus characterized by comprising: The method comprises the steps of: An acquisition module is configured to acquire input data from a target object through a unified search portal, the unified search portal corresponding to a plurality of dimensions of vertical search engines; each vertical search engine is pre-configured with a semantic slot matching the search dimension thereof, the semantic slot comprising at least one slot position; the semantic slot is provided with slot position filling rules; A determination module is configured to perform word segmentation processing on the input data; each word group in the word segmentation result is filled into each slot position matching the slot position type of the semantic slot; wherein the slot position filling rules comprise: if the word segmentation result can fill all slot positions of the semantic slot, it is determined that the search intention of the input data matches the dimension of the vertical search engine corresponding to the semantic slot; or if the word segmentation result can fill a preset number of slot positions in the semantic slot, it is determined that the search intention of the input data matches the dimension of the vertical search engine corresponding to the semantic slot; the search dimension of the vertical search engine corresponding to the semantic slot in which the word segmentation result can fill all slot positions or a preset number of slot positions is determined as the target dimension matching the search intention of the input data; A calling module is configured to determine the word group filled into the slot position of the semantic slot as a search word, and call the vertical search engine of the target dimension to perform a search operation on the search word.

10. An electronic device, comprising: The method comprises the steps of: A processor; A memory for storing processor-executable instructions; The processor executes the executable instructions to implement the method of any one of claims 1-8.

11. A computer readable storage medium having stored thereon computer instructions, wherein, The instructions are executed by the processor to implement the steps of the method of any one of claims 1-8.

Images