Information searching method and device, electronic equipment and storage medium

Abstract

The invention provides an information searching method and device, electronic equipment and a storage medium, and belongs to the technical field of data processing. The information searching method comprises the steps that a first identifier is displayed on an operation interface of an operation and maintenance system; in response to a first operation on the first identifier, receiving to-be-queried information and a selected query mode; determining a target retrieval knowledge base from the full-text retrieval knowledge base and the mixed retrieval knowledge base based on the selected query mode, and performing retrieval in the target retrieval knowledge base based on the to-be-queried information to obtain a retrieval result; and displaying the retrieval result. Inputting the to-be-queried information and the selected query mode by the operation and maintenance personnel through interaction with the first identifier, determining a target retrieval knowledge base from the full-text retrieval knowledge base and the mixed retrieval knowledge base by utilizing the selected query mode at the moment, further performing retrieval in the target retrieval knowledge base by utilizing the to-be-queried information, and displaying a retrieval result. In the process, the searching efficiency is improved while the flexibility of information searching is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to an information searching method and device, electronic equipment and storage medium. BACKGROUND

[0002] The signal system used in railway operation involves a large number of devices, such as signal machines, switch machines, turnouts, on-board devices and various connection lines. The above-mentioned devices have the risk of failure, in order to timely handle the failure of the devices and ensure the normal operation of the signal system, it is necessary to monitor the operating state of the devices in real time, and they are usually managed by an operation and maintenance system.

[0003] The operation and maintenance system is a relatively complex system, and generally requires a developer to provide a user manual for a field worker to read, so that the field worker can smoothly use the intelligent operation and maintenance system.

[0004] However, the way of using the operation manual to assist the field worker in using the operation and maintenance system is not flexible enough. Many times, the worker is not completely clear about the correct reading method of the document, and for a visitor who lacks professional knowledge and wants to experience the operation and maintenance system, it is necessary to refer to the manual while operating, which is relatively low in efficiency, and many times cannot fully meet the needs of the user. SUMMARY

[0005] The present application provides an information searching method and device, electronic equipment and storage medium, to solve the problem that the way of using the operation manual to assist the field worker in using the operation and maintenance system is not flexible enough.

[0006] The present application provides an information searching method, comprising the following steps: displaying a first identifier on an operation interface of an operation and maintenance system, the first identifier being an identifier corresponding to an operation assistant of the operation and maintenance system; in response to a first operation on the first identifier, receiving to-be-searched information and a selected search mode; determining a target search knowledge base from a full-text search knowledge base and a mixed search knowledge base based on the selected search mode, and searching in the target search knowledge base based on the to-be-searched information to obtain a search result; displaying the search result; wherein the full-text search knowledge base and the mixed search knowledge base are knowledge bases constructed based on an operation manual of the operation and maintenance system.

[0007] The information searching method provided by the present application comprises: Based on the selected query mode being the guided question mode, the full-text retrieval knowledge base is used as the target retrieval knowledge base. The full-text retrieval knowledge base includes a hierarchical structure consisting of parent titles and subtitles, with each subtitle associated with a segmented document. The target title that matches the information to be queried is found in the hierarchical structure of the target retrieval knowledge base. Output the segmented documents associated with the target title as the search results.

[0008] The information retrieval method provided by this invention includes, in the step of searching for a target title that matches the information to be queried in the hierarchical structure of the target retrieval knowledge base, the following steps: The target parent title that matches the information to be queried is found in the hierarchical structure of the target retrieval knowledge base. Display all subheadings under the target parent heading; Select the target subheading from all subheadings and use it as the target title.

[0009] The information retrieval method provided by this invention displays all subheadings under the target parent heading in the form of preview images.

[0010] The information retrieval method provided by this invention includes determining a target retrieval knowledge base from a full-text retrieval knowledge base and a hybrid retrieval knowledge base based on the selected query mode, and performing a retrieval in the target retrieval knowledge base based on the information to be queried to obtain retrieval results, including: Based on the selected query mode being a free-question mode, the hybrid retrieval knowledge base is used as the target retrieval knowledge base. The full-text retrieval knowledge base includes multiple embedding vectors corresponding to each segment of the operation manual of the operation and maintenance system. The multiple embedding vectors are vectors output by the embedding model when each segment of the operation manual of the operation and maintenance system is input into the embedding model. The information to be queried is input into a large language model for semantic parsing to obtain the semantic parsing result; Based on the comparison results between the query vector corresponding to the semantic parsing result and the multiple embedding vectors corresponding to each segmented document, the target segmented document is determined; Using the target segmented document as a reference, feedback data corresponding to the information to be queried is generated as the search result.

[0011] The information retrieval method provided by this invention, wherein determining the target segmented document based on the comparison result between the query vector corresponding to the semantic parsing result and multiple embedding vectors corresponding to each segmented document, includes: Obtain the similarity between the query vector corresponding to the semantic parsing result and the multiple embedding vectors corresponding to each segmented document; The segmented documents are sorted in descending order of similarity to obtain the sorting result; The segmented document with the highest similarity in the sorting results is taken as the target segmented document.

[0012] The information retrieval method provided by this invention obtains the full-text retrieval knowledge base and the hybrid retrieval knowledge base through the following methods: Supplementary information is added to each subheading in the operation manual of the operation and maintenance system to obtain the supplemented operation manual of the operation and maintenance system. The supplementary information includes the information of the parent heading corresponding to the subheading. Based on the segmentation information, add segmentation symbols to the supplemented operation and maintenance system operation manual, and divide the supplemented operation and maintenance system operation manual into multiple segmented documents to obtain the updated operation and maintenance system operation manual. The full-text search knowledge base and the hybrid search knowledge base are constructed based on the updated operation manual of the operation and maintenance system. The segmentation information includes the position information of the segmentation symbols supplemented by maintenance personnel.

[0013] This invention provides an information retrieval device, comprising the following modules: The display module is used to display a first identifier on the operation interface of the operation and maintenance system. The first identifier is the identifier corresponding to the operation assistant of the operation and maintenance system. The receiving module is configured to receive the query information and the selected query mode in response to the first operation on the first identifier; The processing module is used to determine the target retrieval knowledge base from the full-text retrieval knowledge base and the hybrid retrieval knowledge base based on the selected query mode, and to perform a retrieval in the target retrieval knowledge base based on the information to be queried, so as to obtain the retrieval results; The display module is also used to display the search results; The full-text search knowledge base and the hybrid search knowledge base are knowledge bases built based on the operation manual of the operation and maintenance system.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the information retrieval method as described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the information retrieval method as described above.

[0016] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the information retrieval method as described above.

[0017] The information retrieval method, apparatus, electronic device, and storage medium provided by this invention display a first identifier on the operation interface of the operation and maintenance system, so that operation and maintenance personnel can enter the information to be queried and the selected query mode through interaction with the first identifier. At this time, the target retrieval knowledge base is determined from the full-text retrieval knowledge base and the hybrid retrieval knowledge base using the selected query mode. Then, the information to be queried is used to search in the target retrieval knowledge base, and the search results are displayed. In this process, the operation and maintenance personnel do not need to manually refer to the operation and maintenance system's operation manual to achieve information retrieval, which improves the flexibility and efficiency of information retrieval. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is one of the flowcharts illustrating the information retrieval method provided by the present invention; Figure 2 This is a flowchart illustrating the process of constructing a full-text retrieval knowledge base and a hybrid retrieval knowledge base provided by the present invention; Figure 3 This is the second flowchart illustrating the information retrieval method provided by the present invention; Figure 4 This is a schematic block diagram of the information retrieval device provided by the present invention; Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] It should be noted that in the description of this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a 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. Unless otherwise specified, 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 that element. The terms "upper," "lower," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly, for example, as a fixed connection, a detachable connection, or an integral connection; a mechanical connection or an electrical connection; a direct connection or an indirect connection through an intermediate medium; or a connection within two elements. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0022] The terms "first," "second," etc., used in this invention are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0023] The following is combined Figures 1-5 The information retrieval method, apparatus, electronic device, and storage medium provided by this invention aim to improve the inflexibility of using operation manuals to assist field staff in using the operation and maintenance system. This allows information retrieval without requiring maintenance personnel to manually consult the operation manual, thereby improving both the flexibility and efficiency of information retrieval.

[0024] Figure 1 This is one of the flowcharts illustrating the information retrieval method provided by the present invention, such as... Figure 1 As shown, including but not limited to the following steps: Step 101: Display the first identifier on the operation interface of the operation and maintenance system. The first identifier is the identifier corresponding to the operation assistant of the operation and maintenance system.

[0025] In some embodiments, the first identifier is a fixed button-type identifier set on the operation interface of the operation and maintenance system.

[0026] For example, a fixed button-style label has the text "Intelligent Operation and Maintenance Fault Handling Assistant".

[0027] Specifically, when maintenance personnel click on the fixed button-like icon, an information input box will appear, whereby the information input box is used to enter the information to be queried.

[0028] Step 102: In response to the first operation on the first identifier, receive the query information and the selected query mode.

[0029] In some embodiments, the first operation may be a basic click operation. For example, in the "Signal Parameter Configuration" interface, which is one of the operation interfaces of the operation and maintenance system, the operation and maintenance personnel left-click the fixed button with the text "Intelligent Operation and Maintenance Fault Handling Assistant" to trigger the assistant pop-up window, which is the information input box in this invention. The assistant pop-up window displays "Please enter the information to be queried".

[0030] In some embodiments, the information to be queried entered in the information input box is selected by default as guided question mode. That is, when the maintenance personnel enter the information to be queried in the information input box and trigger the search function, the selected query mode is guided question mode by default.

[0031] In some embodiments, when the characters "free to ask questions" are entered in the information input box, the selected query mode is switched from the default guided question mode to the free question mode, and the information entered again in the information input box is used as the query information.

[0032] Step 103: Based on the selected query mode, determine the target retrieval knowledge base from the full-text retrieval knowledge base and the hybrid retrieval knowledge base, and perform a retrieval in the target retrieval knowledge base based on the information to be queried to obtain the retrieval results.

[0033] Step 104: Display the search results.

[0034] Among them, the full-text search knowledge base and the hybrid search knowledge base are knowledge bases built based on the operation and maintenance system's operation manual.

[0035] In this embodiment, a first identifier is displayed on the operation interface of the operation and maintenance system, so that operation and maintenance personnel can enter the information to be queried and the selected query mode through interaction with the first identifier. At this time, the target retrieval knowledge base is determined from the full-text retrieval knowledge base and the hybrid retrieval knowledge base using the selected query mode. Then, the information to be queried is used to search in the target retrieval knowledge base, and the search results are displayed. In this process, the operation and maintenance personnel do not need to manually refer to the operation manual of the operation and maintenance system to find information, which improves the flexibility of information retrieval and the efficiency of retrieval.

[0036] In some embodiments, a target retrieval knowledge base is determined from a full-text retrieval knowledge base and a hybrid retrieval knowledge base based on a selected query mode, and a retrieval is performed in the target retrieval knowledge base based on the information to be queried to obtain retrieval results, including: The selection-based query mode is a guided question mode that uses the full-text search knowledge base as the target search knowledge base. The full-text search knowledge base includes a hierarchical structure consisting of parent headings and subheadings, with each subheading associated with a segmented document. Search the hierarchical structure of the target retrieval knowledge base for target titles that match the information to be queried; Output the segmented document associated with the target title as the search results.

[0037] In the operation manuals of traditional operation and maintenance systems, the operation instructions for a certain type of equipment may be mixed with other content (such as principle descriptions and historical fault records), and maintenance personnel need to sift through large blocks of text to find the key steps.

[0038] In this embodiment, the hierarchical structure of the full-text retrieval knowledge base, consisting of parent and subheadings, can be used to accurately locate the target title that matches the information to be queried. This eliminates the need for maintenance personnel to conduct extensive searches, allowing them to locate the target title that matches the information to be queried and obtain the segmented document associated with the target title.

[0039] For highly standardized railway operations, such as turnout fault handling which must follow the sequence of "checking the power supply → inspecting mechanical components → restarting the equipment," the standardized procedures can be directly embedded into the search results. This ensures that maintenance personnel receive guidance based on "industry-standard operations," avoiding non-standard operations caused by misinterpretations of manuals and reducing equipment damage or operational risks.

[0040] In some embodiments, a segmented document can be understood as a "minimum information unit" formed by splitting the operation and maintenance system manual after preprocessing. It has independent information value and a clear ownership. It is neither a large block of text in the complete manual nor a scattered short sentence, but a structured text fragment that "can independently answer a specific question and correspond to a specific operation scenario".

[0041] Furthermore, the guided question-and-answer search method can be adapted to the differentiated needs of different user identities. Specifically, for new maintenance personnel, the hierarchical menu of guided questions acts like a "navigation," helping them pinpoint their needs step by step, coupled with precise segmented documents for quick onboarding. For visitors, there is no need to understand technical jargon; they can directly obtain simplified segmented documents (such as the meaning of traffic signal light colors) through menu selection (e.g., "Traffic Signal → Function Introduction"), meeting the needs of "experiencing the system" and aligning with the original intention of the handover document to "meet the needs of different users."

[0042] In some embodiments, searching for a target title that matches the information to be queried within the hierarchical structure of the target retrieval knowledge base includes: Search the hierarchical structure of the target retrieval knowledge base for the target parent title that matches the information to be queried; Display all subheadings under the target parent heading; Select the target subheading from all subheadings and use it as the target title.

[0043] In this embodiment, the three-layer retrieval logic of "first find the target parent title → display the subtitle → then find the target subtitle" can lower the selection threshold, avoid the problems of "not being able to select accurately or not being able to find", and make the search for the target title smoother.

[0044] Specifically, the hierarchical structure of the operation and maintenance system is complex (e.g., the parent title "Signal Machine" may contain multiple subtitles such as "Lighting Fault," "Parameter Calibration," and "Routine Maintenance"). If users are asked to directly input the complete title at once (e.g., "Signal Machine - Lighting Fault - Red Light Not Working"), new operation and maintenance personnel or non-professional users are prone to confusing the hierarchy and entering incorrect content. However, according to this hierarchical logic, users only need to first determine the parent title "Signal Machine," and the system will automatically list all its subtitles (e.g., "Lighting Fault," "Parameter Calibration," etc.). Users can intuitively select from these, essentially providing a "visual navigation menu," eliminating the need to memorize the complete path and significantly lowering the operational threshold.

[0045] In addition, the above-mentioned search method, as a step-by-step search method, reduces the efficiency loss caused by selecting the wrong title at once, while also achieving accurate positioning of the target title.

[0046] Specifically, by finding the target parent title that matches the information to be queried, the search scope can be narrowed, reducing "irrelevant matches" at the source. By utilizing the user's secondary filtering to find the target title under the parent title, the search scope is narrowed from the "entire knowledge base" to the "subtitles under the parent title", ensuring that the search direction does not go astray, thereby improving the accuracy of the search results.

[0047] In some embodiments, all subheadings under the target parent heading are displayed as preview images.

[0048] In this embodiment, by displaying all subheadings under the target parent heading in the form of preview images, maintenance personnel can view all subheadings more comprehensively and then select the subheading that the user wants to view in order to meet the needs of maintenance personnel.

[0049] In some embodiments, there can be multiple parent titles, each parent title corresponds to multiple subtitles, and each parent title can display all subtitles in the form of a preview image for easy reference by maintenance personnel.

[0050] For example, the parent title can be understood as a menu, and the subtitle can be understood as a submenu contained under the menu. At this time, maintenance personnel can select the submenu displayed in the form of a preview image as needed, so that users can select the tutorial of a certain module they want to view by referring to the operation and maintenance system interface. After determining the interface they want to view, they can return to the operation guide corresponding to that interface / module / menu.

[0051] In some embodiments, a target retrieval knowledge base is determined from a full-text retrieval knowledge base and a hybrid retrieval knowledge base based on a selected query mode, and a retrieval is performed in the target retrieval knowledge base based on the information to be queried to obtain retrieval results, including: The selected query mode is a free-question mode, and the hybrid retrieval knowledge base is used as the target retrieval knowledge base. The full-text retrieval knowledge base includes multiple embedding vectors corresponding to each segment of the operation manual of the operation and maintenance system. The multiple embedding vectors are the vectors output by the embedding model when each segment of the operation manual of the operation and maintenance system is input into the embedding model. The information to be queried is input into a large language model for semantic parsing, and the semantic parsing results are obtained. The target segment document is determined by comparing the query vector corresponding to the semantic parsing result with the multiple embedding vectors corresponding to each segment document. Using the target segmented document as a reference, generate feedback data corresponding to the information to be queried, which will serve as the search results.

[0052] In this embodiment, experienced operations and maintenance personnel or those with vague needs can use the free-question mode to perform searches, completely breaking free from the "menu constraints" and obtaining information more freely and efficiently.

[0053] For example, in railway operation and maintenance scenarios, maintenance personnel often have "clear needs but flexible expressions," such as wanting to search for "what to do if the maintenance machine cannot connect to the turnout," rather than strictly following the menu to select "maintenance machine → connection fault → turnout connection." If relying on traditional keyword searches, the search may fail due to "inconsistent expression," such as omitting keywords like "connection" or "fault."

[0054] In this embodiment, the information to be queried is input into a large language model for semantic parsing to obtain the semantic parsing results. This allows for the inference of the maintenance personnel's intent, and then the target segmented document is selected from multiple segmented documents based on the query vector corresponding to the semantic parsing results. In this process, the problem of flexible expression but inaccurate retrieval can be avoided.

[0055] By using the query vector corresponding to the semantic parsing result to filter out the target segment document from multiple segmented documents, the target segment document that is closest to the query vector corresponding to the semantic parsing result can be selected. Obviously, in this process, the most suitable target segment document can be selected. Feedback is generated based on the target segment document, which is equivalent to adding a "precise anchor" to semantic retrieval. It retains the flexibility of semantic retrieval and avoids the retrieval results from going astray through the "semantic anchoring" of the embedded vector, achieving the effect of "flexible without going astray and precise without being rigid".

[0056] In some embodiments, the target segmented document is determined based on the comparison results between the query vector corresponding to the semantic parsing result and multiple embedding vectors corresponding to each segmented document, including: Obtain the similarity between the query vector corresponding to the semantic parsing result and the multiple embedding vectors corresponding to each segmented document; The multiple segmented documents are sorted from highest to lowest similarity to obtain the sorting results; The segmented document with the highest similarity in the ranking results is used as the target segmented document.

[0057] In this embodiment, similarity is calculated to sort multiple segmented documents. During this process, the segmented document with the highest similarity in the sorted results can be selected as the target segmented document.

[0058] Specifically, if maintenance personnel use flexible expressions, such as "the maintenance machine cannot connect to the switch" or "the communication between the switch and the maintenance machine is lost," relying solely on keywords such as "cannot connect" or "communication is lost" for matching may result in missing core documents due to differences in vocabulary. However, this invention calculates the similarity between the "query vector" and the "embedded vector," which essentially compares the "distance" between the "question semantics" and the "document semantics." As long as the core semantics of the two are consistent, even if the expressions are different, the similarity will be high, solving the pain point of "flexible expressions but missed items in retrieval" and ensuring that the most relevant documents are not missed.

[0059] By selecting the segmented document with the highest similarity as the target segmented document, the content with the highest overlap with the user's needs can be matched, avoiding interference from multiple results and obtaining core information without secondary filtering.

[0060] In some embodiments, the full-text search knowledge base and the hybrid search knowledge base are obtained through the following methods: Add supplementary information to each subheading in the operation and maintenance system's operation manual to obtain the supplemented operation and maintenance system's operation manual. The supplementary information includes the information of the parent heading corresponding to the subheading. Based on the segmentation information, add segmentation symbols to the supplemented operation and maintenance system operation manual, and divide the supplemented operation and maintenance system operation manual into multiple segmented documents to obtain the updated operation and maintenance system operation manual. A full-text search knowledge base and a hybrid search knowledge base are built based on the updated operation and maintenance system's user manual; The segmentation information includes the location information of the segmentation symbols supplemented by maintenance personnel.

[0061] In this embodiment, by adding supplementary information, a parent-child relationship is established between different headings in the operation and maintenance system's manual, and a full-text retrieval knowledge base is constructed based on this.

[0062] Meanwhile, by using the location information of the segmentation symbols supplemented by maintenance personnel to add segmentation symbols to the supplemented operation and maintenance system's operation manual, the operation and maintenance system's operation manual can be divided into multiple segmented documents. In this process, the introduction of segmentation information can ensure that the segmented documents obtained are reasonable, and ensure the accuracy of the full-text search knowledge base and the hybrid search knowledge base.

[0063] In some embodiments, the Dify program is used to build a full-text search knowledge base and a hybrid search knowledge base. The Dify program is an open-source large language model application development platform, which will not be described in detail here.

[0064] Specifically, such as Figure 2 As shown, the operation manual of the operation and maintenance system is first preprocessed. The preprocessing includes: supplementing the headings at all levels, adding the parent heading information to each subheading so that maintenance personnel can locate the relevant content more accurately, and manually splitting each organic segment by adding obvious separators between segments to increase the accuracy of segmentation when building the knowledge base.

[0065] Secondly, import the pre-processed operation manual of the operation and maintenance system into the knowledge base, divide it into segments using the segmentation method mentioned in the previous step, and select two retrieval modes to build the knowledge bases respectively, resulting in a full-text retrieval knowledge base and a hybrid retrieval knowledge base.

[0066] The full-text search knowledge base provides a guided questioning process, enabling users to accurately match the table content they want to query. The hybrid search knowledge base takes into account both the accuracy of sentence matching and semantic commonality, and provides a free questioning mode, enabling the embedded model to better obtain the answers users want from the knowledge base. In addition, it sets search thresholds to filter out segments with low matching degree, reducing the possibility of users getting irrelevant answers.

[0067] In some embodiments, the guided question mode provides users with clear menu guidance, giving them a clear direction for their search and enabling them to accurately determine the specific content they want to query. In contrast, the free-question mode allows users to determine their own query content, and the workflow matches relevant information from the knowledge base and presents it to the user.

[0068] For example, such as Figure 3 As shown, information retrieval methods include: Step 301, input by maintenance personnel.

[0069] Step 302, guided questioning mode.

[0070] Step 303: Extract the first-level menu information from the input of the maintenance personnel.

[0071] Step 304: Query all second-level menus under this first-level menu.

[0072] Step 305: Maintenance personnel select the corresponding secondary menu.

[0073] Step 306: The workflow extracts the content under the secondary menu from the knowledge base.

[0074] Step 307, Open Question Mode.

[0075] Step 308: Query the corresponding content in the knowledge base based on the input of the maintenance personnel.

[0076] Step 309: The workflow organizes the retrieved content into language and returns it to the maintenance personnel.

[0077] It should be noted that the information retrieval device provided by the present invention can execute the information retrieval method of any of the above embodiments during specific operation, which will not be elaborated in this embodiment.

[0078] like Figure 4 As shown, the information retrieval device provided by the present invention includes: Display module 401 is used to display a first identifier on the operation interface of the operation and maintenance system. The first identifier is the identifier corresponding to the operation assistant of the operation and maintenance system. The receiving module 402 is configured to receive the query information and the selected query mode in response to the first operation on the first identifier; The processing module 403 is used to determine the target retrieval knowledge base from the full-text retrieval knowledge base and the hybrid retrieval knowledge base based on the selected query mode, and to perform a retrieval in the target retrieval knowledge base based on the information to be queried, so as to obtain the retrieval results; Display module 401 is also used to display search results; Among them, the full-text search knowledge base and the hybrid search knowledge base are knowledge bases built based on the operation and maintenance system's operation manual.

[0079] In this embodiment, a first identifier is displayed on the operation interface of the operation and maintenance system, so that operation and maintenance personnel can enter the information to be queried and the selected query mode through interaction with the first identifier. At this time, the target retrieval knowledge base is determined from the full-text retrieval knowledge base and the hybrid retrieval knowledge base using the selected query mode. Then, the information to be queried is used to search in the target retrieval knowledge base, and the search results are displayed. In this process, the operation and maintenance personnel do not need to manually refer to the operation manual of the operation and maintenance system to find information, which improves the flexibility of information retrieval and the efficiency of retrieval.

[0080] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call logical instructions in the memory 530 to execute an information retrieval method, which includes: displaying a first identifier on the operation interface of the operation and maintenance system, the first identifier being an identifier corresponding to the operation assistant of the operation and maintenance system; responding to a first operation on the first identifier, receiving information to be queried and a selected query mode; determining a target retrieval knowledge base from the full-text retrieval knowledge base and the hybrid retrieval knowledge base based on the selected query mode, and performing a retrieval in the target retrieval knowledge base based on the information to be queried to obtain retrieval results; and displaying the retrieval results; wherein the full-text retrieval knowledge base and the hybrid retrieval knowledge base are knowledge bases constructed based on the operation manual of the operation and maintenance system.

[0081] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0082] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer is able to execute the information retrieval method provided in the above embodiments. The method includes: displaying a first identifier on the operation interface of the operation and maintenance system, the first identifier being an identifier corresponding to the operation assistant of the operation and maintenance system; receiving information to be queried and a selected query mode in response to a first operation on the first identifier; determining a target retrieval knowledge base from a full-text retrieval knowledge base and a hybrid retrieval knowledge base based on the selected query mode, and performing a retrieval in the target retrieval knowledge base based on the information to be queried to obtain retrieval results; and displaying the retrieval results; wherein the full-text retrieval knowledge base and the hybrid retrieval knowledge base are knowledge bases constructed based on the operation manual of the operation and maintenance system.

[0083] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the information retrieval method provided in the above embodiments. The method includes: displaying a first identifier on the operation interface of the operation and maintenance system, the first identifier being an identifier corresponding to the operation assistant of the operation and maintenance system; receiving information to be queried and a selected query mode in response to a first operation on the first identifier; determining a target retrieval knowledge base from a full-text retrieval knowledge base and a hybrid retrieval knowledge base based on the selected query mode, and performing a retrieval in the target retrieval knowledge base based on the information to be queried to obtain retrieval results; and displaying the retrieval results; wherein the full-text retrieval knowledge base and the hybrid retrieval knowledge base are knowledge bases constructed based on the operation manual of the operation and maintenance system.

[0084] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0085] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An information retrieval method, characterized in that, include: The first identifier is displayed on the operation interface of the operation and maintenance system. The first identifier is the identifier corresponding to the operation assistant of the operation and maintenance system. In response to the first operation on the first identifier, receive the information to be queried and the selected query mode; Based on the selected query mode, a target retrieval knowledge base is determined from the full-text retrieval knowledge base and the hybrid retrieval knowledge base, and a retrieval is performed on the target retrieval knowledge base based on the information to be queried to obtain the retrieval results; Display the search results; The full-text search knowledge base and the hybrid search knowledge base are knowledge bases built based on the operation manual of the operation and maintenance system.

2. The information retrieval method according to claim 1, characterized in that, The method involves determining a target retrieval knowledge base from the full-text retrieval knowledge base and the hybrid retrieval knowledge base based on the selected query mode, and then performing a search on the target retrieval knowledge base based on the information to be queried to obtain search results, including: Based on the selected query mode being the guided question mode, the full-text retrieval knowledge base is used as the target retrieval knowledge base. The full-text retrieval knowledge base includes a hierarchical structure consisting of parent titles and subtitles, with each subtitle associated with a segmented document. The target title that matches the information to be queried is found in the hierarchical structure of the target retrieval knowledge base. Output the segmented documents associated with the target title as the search results.

3. The information retrieval method according to claim 2, characterized in that, The step of searching for a target title that matches the information to be queried within the hierarchical structure of the target retrieval knowledge base includes: The target parent title that matches the information to be queried is found in the hierarchical structure of the target retrieval knowledge base. Display all subheadings under the target parent heading; Select the target subheading from all subheadings and use it as the target title.

4. The information retrieval method according to claim 3, characterized in that, All subheadings under the target parent heading are displayed in preview form.

5. The information retrieval method according to claim 1, characterized in that, The method involves determining a target retrieval knowledge base from the full-text retrieval knowledge base and the hybrid retrieval knowledge base based on the selected query mode, and then performing a search on the target retrieval knowledge base based on the information to be queried to obtain search results, including: Based on the selected query mode being a free-question mode, the hybrid retrieval knowledge base is used as the target retrieval knowledge base. The full-text retrieval knowledge base includes multiple embedding vectors corresponding to each segment of the operation manual of the operation and maintenance system. The multiple embedding vectors are vectors output by the embedding model when each segment of the operation manual of the operation and maintenance system is input into the embedding model. The query information is input into a large language model for semantic parsing to obtain the semantic parsing result; Based on the comparison results between the query vector corresponding to the semantic parsing result and the multiple embedding vectors corresponding to each segmented document, the target segmented document is determined; Using the target segmented document as a reference, feedback data corresponding to the information to be queried is generated as the search result.

6. The information retrieval method according to claim 5, characterized in that, The step of determining the target segmented document based on the comparison results between the query vector corresponding to the semantic parsing result and the multiple embedding vectors corresponding to each segmented document includes: Obtain the similarity between the query vector corresponding to the semantic parsing result and the multiple embedding vectors corresponding to each segmented document; The segmented documents are sorted in descending order of similarity to obtain the sorting result; The segmented document with the highest similarity in the sorting results is taken as the target segmented document.

7. The information retrieval method according to any one of claims 1 to 6, characterized in that, The full-text search knowledge base and the hybrid search knowledge base were obtained through the following methods: Supplementary information is added to each subheading in the operation manual of the operation and maintenance system to obtain the supplemented operation manual of the operation and maintenance system. The supplementary information includes the information of the parent heading corresponding to the subheading. Based on the segmentation information, add segmentation symbols to the supplemented operation and maintenance system operation manual, and divide the supplemented operation and maintenance system operation manual into multiple segmented documents to obtain the updated operation and maintenance system operation manual. The full-text search knowledge base and the hybrid search knowledge base are constructed based on the updated operation manual of the operation and maintenance system. The segmentation information includes the position information of the segmentation symbols supplemented by maintenance personnel.

8. An information retrieval device, characterized in that, include: The display module is used to display a first identifier on the operation interface of the operation and maintenance system. The first identifier is the identifier corresponding to the operation assistant of the operation and maintenance system. The receiving module is configured to receive the query information and the selected query mode in response to the first operation on the first identifier; The processing module is used to determine the target retrieval knowledge base from the full-text retrieval knowledge base and the hybrid retrieval knowledge base based on the selected query mode, and to perform a retrieval in the target retrieval knowledge base based on the information to be queried, so as to obtain the retrieval results; The display module is also used to display the search results; The full-text search knowledge base and the hybrid search knowledge base are knowledge bases built based on the operation manual of the operation and maintenance system.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the information retrieval method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the information retrieval method as described in any one of claims 1 to 7.

Images