Database scene-oriented information retrieval method and device and electronic equipment
By receiving recall requests in a distributed system, performing information recalls on different data nodes, and optimizing the ranking of candidate information based on node ranking, the problem of insufficient local sorting accuracy is solved, and efficient information retrieval without increasing memory and network consumption is achieved.
Patent Information
- Application Number
- CN202411979637.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-30
- Publication Date
- 2025-05-06
AI Technical Summary
In distributed systems, local sorting is not as accurate as global sorting, and there are randomness problems, while global sorting requires huge amounts of memory and network consumption, and has high time complexity.
By receiving a recall request, information recall is performed on different data nodes, candidate information is obtained, and candidate information is optimized according to the node ranking, and finally feedback target information to the client.
Without increasing memory and network consumption, the accuracy and efficiency of information retrieval are improved, and the ranking of data nodes is predicted through the distribution of historical recall information, thereby optimizing the ranking of search results.
Smart Images
Figure CN119938722A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of data processing technology, specifically to the field of information recommendation technology, and in particular to an information retrieval method, device and electronic device for database scenarios. Background Art
[0002] In distributed systems, local ranking is usually used instead of global ranking for network and performance reasons to reduce memory and network consumption. However, local ranking is not as accurate as global ranking and has a large randomness problem; global ranking requires a huge amount of memory and network consumption, and has a high time complexity. Summary of the invention
[0003] The present disclosure provides an information retrieval method, device and electronic device for database scenarios.
[0004] According to one aspect of the present disclosure, there is provided an information retrieval method for a database scenario, comprising: receiving a recall request, and performing information recall at different data nodes according to the recall request to obtain at least one candidate information to be recalled; obtaining node rankings of different data nodes; optimizing the ranking of the candidate information based on the node rankings, and feeding back target information to a client according to the optimized ranking of the candidate information.
[0005] According to another aspect of the present disclosure, there is provided an information retrieval device for a database scenario, comprising: a recall module for receiving a recall request, and performing information recall at different data nodes according to the recall request to obtain at least one candidate information to be recalled; an acquisition module for obtaining node rankings of different data nodes; an optimization module for optimizing the ranking of the candidate information based on the node ranking, and feeding back target information to the client according to the optimized ranking of the candidate information.
[0006] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the information retrieval method for database scenarios described in the above-mentioned first aspect embodiment.
[0007] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, on which a computer program / instructions are stored. The computer instructions are used to enable the computer to execute the information retrieval method for database scenarios described in the above-mentioned embodiment.
[0008] According to another aspect of the present disclosure, a computer program product is provided, including a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the information retrieval method for a database scenario described in the above-mentioned first embodiment is implemented.
[0009] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.
[0011] Figure 1 A flowchart of an information retrieval method for a database scenario provided by an embodiment of the present disclosure;
[0012] Figure 2 A flowchart of another information retrieval method for a database scenario provided by an embodiment of the present disclosure;
[0013] Figure 3 A flowchart of another information retrieval method for a database scenario provided by an embodiment of the present disclosure;
[0014] Figure 4 A schematic diagram of the structure of an information retrieval device for database scenarios provided by an embodiment of the present disclosure;
[0015] Figure 5 The present invention is a block diagram of an electronic device for implementing the information retrieval method for database scenarios according to an embodiment of the present disclosure. DETAILED DESCRIPTION
[0016] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.
[0017] The following describes the information retrieval method, device and electronic device for database scenarios according to the embodiments of the present disclosure with reference to the accompanying drawings.
[0018] Figure 1 The following is a flowchart of an information retrieval method for database scenarios provided in an embodiment of the present application. Figure 1 As shown, the method may include but is not limited to the following steps:
[0019] S101, receiving a recall request.
[0020] In some embodiments, a recall request sent by a client may be received.
[0021] In some embodiments, the recall request may include relevant search terms for the desired information.
[0022] In some embodiments, the related search terms included in the recall request may include multi-dimensional search terms.
[0023] In some embodiments, the client may send a recall request to the search engine. Accordingly, the search engine may receive the recall request sent by the client, and then perform at least one information search on different data nodes according to the recall request.
[0024] For example, the recall request is "beautiful, pure cotton purple dress". It can be understood that the words "beautiful", "purple" and "pure cotton" in the recall request are related search terms in different dimensions.
[0025] S102: Recall information at different data nodes according to the recall request to obtain at least one candidate information to be recalled.
[0026] In some embodiments, in distributed information storage, different information can be stored on different data nodes. For example, taking clothes as an example, information about the first batch of clothes can be stored on data node A, information about the second batch of clothes can be stored on data node B, and information about the third batch of clothes can be stored on data node C.
[0027] Information recall can be performed on different data nodes based on the recall request. Optionally, information recall in multiple dimensions can be performed on different data nodes based on the recall request to obtain at least one candidate information to be recalled. It is understandable that the recalled candidate information can come from each data node, one data node, or some of all data nodes.
[0028] For example, if the recall request is "beautiful, pure cotton purple dress", multiple dimensions of dresses can be searched on different data nodes based on search terms such as "beautiful", "purple" and "pure cotton". Each data node can be searched according to the relevant search terms, and ultimately multiple candidate dresses can be returned from multiple data nodes.
[0029] S103, obtaining node rankings of different data nodes.
[0030] In some embodiments, historical recall information corresponding to historical recall requests may be obtained, and further, node rankings of different data nodes may be determined based on the historical recall information.
[0031] In some embodiments, the distribution of historical recall information on different data nodes is determined, and the data nodes are further ranked according to the distribution of the data nodes to obtain node rankings of different data nodes.
[0032] For example, the data nodes include data node A, data node B, data node C, and data node D. Among the 100,000 historical recall requests, 50% of the historical recall information comes from data node A, 30% of the recall information comes from data node B, and 20% of the recall information comes from data node D. Based on the distribution of these historical recall information, the node rankings of the data nodes can be determined as data node A, data node B, data node D, and data node C.
[0033] S104, optimizing the ranking of the candidate information based on the node ranking, and feeding back the target information to the client according to the optimized ranking of the candidate information.
[0034] In some embodiments, after obtaining the candidate information, the ranking optimization parameters of the data node where the candidate information is located can be determined based on the node ranking, and the ranking of the candidate information recalled by the data node can be optimized according to the ranking optimization parameters of the data node. Furthermore, the target information can be fed back to the client based on the optimized ranking of the candidate information.
[0035] In some embodiments, the candidate information is sorted in descending order according to the optimized ranking, part of the candidate information is selected as the target information according to the sorting result, and the target information is fed back to the client.
[0036] In some embodiments, candidate information ranked in the top K is selected from the candidate information as target information, and the target information is fed back to the client, where the value of K is an integer greater than or equal to 1.
[0037] In an embodiment of the present application, the distribution of historical recall information on data nodes can be used to predict the ranking of data nodes. Furthermore, based on the ranking of data nodes, the ranking of searched information can be optimized or adjusted, thereby using the ranking of data nodes as a reference factor for the global rank. This can increase global reference factors in local multi-way searches, thereby achieving the goal of balancing memory and network consumption. That is, the accuracy of search results can be enhanced without increasing memory and network consumption.
[0038] Figure 2 The following is a flowchart of an information retrieval method for database scenarios provided in an embodiment of the present application. Figure 2 As shown, the method may include but is not limited to the following steps:
[0039] S201, receiving a recall request.
[0040] S202: Recall information at different data nodes according to the recall request to obtain at least one candidate information to be recalled.
[0041] For a detailed description of steps S201 to S202, please refer to the relevant contents in the above embodiment, which will not be repeated here.
[0042] S203: Obtain historical recall information corresponding to the historical recall request.
[0043] S204: Determine the node rankings of different data nodes according to the historical recall information.
[0044] In some embodiments, the distribution of historical recall information on different data nodes is determined, and further, the data nodes are ranked according to the distribution of the data nodes to obtain node rankings of different data nodes.
[0045] In some embodiments, a first total amount of historical recalled information is determined, and further, based on the distribution of the data nodes, a second total amount of recalled information of the data nodes is determined. Further, based on the first total amount of information and the second total amount of information, the data nodes are sorted to obtain node rankings of different data nodes.
[0046] In some embodiments, the proportion of recalled information corresponding to the data node is determined based on the total amount of the first information and the total amount of the second information, and further, the data nodes are sorted based on the proportion of recalled information to obtain node rankings of different data nodes. The ranking of the data node can be determined in the above manner, the calculation method is relatively simple, and more computing resources are not used, which improves the speed of determining the ranking of the data node.
[0047] In some embodiments, according to the distribution of data nodes, the historical information ranking of the historical recall information on the data nodes is determined, and the historical information ranking is averaged or weighted to obtain the node ranking of the data node.
[0048] Optionally, the historical information ranking of the historical recall information recalled by each historical recall request is determined, and the historical recall information belonging to the same data node is determined, and the historical information rankings of all historical recall information on the data node are averaged to obtain the information ranking mean of the historical recall information, and the information ranking mean is used as the node ranking of the data node.
[0049] Optionally, the historical information ranking of the historical recall information recalled by each historical recall request is determined, and the historical recall information belonging to the same data node is determined, and the historical information ranking of all the historical recall information on the data node is weighted to obtain a weighted information ranking, and the weighted information ranking is used as the node ranking of the data node.
[0050] By weighting or averaging the historical information ranking of the historical recall information, the node ranking of the data node is made closer to the information ranking of the historical recall information, so that the node ranking of the data node can better reflect the ranking of the information.
[0051] In some embodiments, a pre-trained ranking prediction model can be called, and the distribution of data nodes can be input into the ranking prediction model, and the ranking prediction model can output the node rankings of different data nodes. The node rankings of data nodes are predicted by the artificial intelligence model, which integrates the advantages of artificial intelligence and can make the node rankings of data nodes more accurate.
[0052] In an embodiment of the present application, the distribution of historical recall information on data nodes can be used to predict the ranking of data nodes. Through the ranking of data nodes, global reference factors can be added to local multi-way searches, thereby reflecting the global ranking of search results.
[0053] S205, optimizing the ranking of the candidate information based on the node ranking, and feeding back the target information to the client according to the optimized ranking of the candidate information.
[0054] In an embodiment of the present application, the distribution of historical recall information on data nodes can be used to predict the ranking of data nodes. Furthermore, based on the ranking of data nodes, the ranking of searched information can be optimized or adjusted. Since the rankings of shards on different indexes are different, the ranking of data nodes is used as a reference factor for the global rank. Therefore, the ranking of data nodes is used as a reference factor for the global rank. This can increase the global reference factor in local multi-way search and achieve the purpose of balancing memory and network consumption. That is, the accuracy of search results can be enhanced without increasing memory and network consumption.
[0055] Figure 3 The following is a flowchart of an information retrieval method for database scenarios provided in an embodiment of the present application. Figure 3 As shown, the method may include but is not limited to the following steps:
[0056] S301, receiving a recall request.
[0057] S302: Recall information at different data nodes according to the recall request to obtain at least one candidate information to be recalled.
[0058] S303: Obtain historical recall information corresponding to the historical recall request.
[0059] S304: Determine the node rankings of different data nodes according to the historical recall information.
[0060] S305: Determine at least one target data node to which the candidate information belongs, and an initial first information ranking of the candidate information on the target data node.
[0061] For a detailed description of steps S301 to S205, please refer to the relevant contents in the above embodiment, which will not be repeated here.
[0062] S306: Optimize the first information ranking according to the node ranking of the target data node to obtain a second information ranking of the candidate information on the target data node.
[0063] S307: Obtain a global optimized ranking of the candidate information according to the second information ranking of the candidate information on the target data node.
[0064] In some embodiments, a ranking optimization parameter corresponding to the target data node is determined based on the node ranking of the target data node, and further, the first information ranking is optimized based on the ranking optimization parameter to obtain a second information ranking of the candidate information on the target data node.
[0065] In an embodiment of the present application, the ranking of the searched information can be optimized or adjusted based on the ranking of the data nodes, that is, the global reference factors can be added to the local multi-way search. Since the optimization is only performed through the ranking of the data nodes, the optimized ranking of the candidate information can be obtained without causing a significant increase in memory and network consumption relative to the global rank.
[0066] In some embodiments, the following formula may be used to determine the first information ranking of the candidate information on the target data node:
[0067]
[0068] R is a set, representing a collection of certain elements or resources related to the candidate information d.
[0069] k is a constant used to adjust the weight of the ranking.
[0070] Rank(d) represents the position of candidate information d in a certain sorting.
[0071] In some embodiments, the following formula may be used to determine the second information ranking of the candidate information on the target data node:
[0072]
[0073] Wherein, epsilon is a ranking optimization parameter calculated according to the node ranking of the data node. Optionally, the ranking optimization parameter may be an amplification parameter.
[0074] In some embodiments, the node ranking of the target data node is mapped to determine a mapping value within a target numerical range, and further, the mapping value is determined as a ranking adjustment parameter corresponding to the target data node.
[0075] In some embodiments, the following formula may be used to determine the ranking optimization parameter of the target data node:
[0076]
[0077] Where x is the node ranking of the target data node. The above formula can map epsilon to between (1 and 1.5), so that the ranking optimization parameters converge within this interval, and the optimization of the first information ranking will not deviate greatly, which can ensure the accuracy of the information ranking. When x is larger, epsilon is larger and RRF(d) is smaller, thus ensuring a higher ranking score on the hot data node.
[0078] S308: Feedback target information to the client based on the optimized ranking of the candidate information.
[0079] For a detailed description of step S308, please refer to the relevant contents in the above embodiment, which will not be repeated here.
[0080] In an embodiment of the present application, the distribution of historical recall information on data nodes can be used to predict the ranking of data nodes. Furthermore, based on the ranking of data nodes, the ranking of searched information can be optimized or adjusted. Since the rankings of data nodes on different indexes are different, the ranking of data nodes is used as a reference factor for the global rank. This can increase the global reference factor in local multi-way search and achieve the purpose of balancing memory and network consumption. That is, the accuracy of search results can be enhanced without increasing memory and network consumption.
[0081] Corresponding to the information retrieval methods for database scenarios provided in the above-mentioned embodiments, an embodiment of the present disclosure also provides an information retrieval device for database scenarios. Since the information retrieval device for database scenarios provided in the embodiment of the present disclosure corresponds to the information retrieval methods for database scenarios provided in the above-mentioned embodiments, the implementation methods of the above-mentioned information retrieval methods for database scenarios are also applicable to the information retrieval device for database scenarios provided in the embodiment of the present disclosure, and will not be described in detail in the following embodiments.
[0082] Figure 4 A schematic diagram of the structure of an information retrieval device for database scenarios provided in an embodiment of the present disclosure.
[0083] like Figure 4 As shown, the information retrieval device 400 for database scenarios according to an embodiment of the present disclosure includes a recall module 401 , an acquisition module 402 and an optimization module 403 .
[0084] The recall module 401 is used to receive a recall request and perform information recall at different data nodes according to the recall request to obtain at least one candidate information to be recalled;
[0085] An acquisition module 402 is used to acquire node rankings of different data nodes;
[0086] The optimization module 403 is used to optimize the ranking of the candidate information based on the node ranking, and feed back target information to the client according to the optimized ranking of the candidate information.
[0087] In one embodiment of the present disclosure, the acquisition module 402 is further used to: acquire historical recall information corresponding to the historical recall request; and determine the node rankings of the different data nodes according to the historical recall information.
[0088] In one embodiment of the present disclosure, the acquisition module 402 is further used to: determine the distribution of the historical recall information on the different data nodes; and rank the data nodes according to the distribution of the data nodes to obtain the node rankings of the different data nodes.
[0089] In one embodiment of the present disclosure, the acquisition module 402 is also used to: determine a first total amount of information of the historical recall information; determine a second total amount of information of the data node being recalled based on the distribution information of the data node; sort the data nodes based on the first total amount of information and the second total amount of information to obtain node rankings of the different data nodes.
[0090] In one embodiment of the present disclosure, the acquisition module 402 is also used to: determine the proportion of recalled information corresponding to the data node based on the first total amount of information and the second total amount of information; sort the data nodes according to the proportion of recalled information to obtain the node rankings of the different data nodes.
[0091] In one embodiment of the present disclosure, the acquisition module 402 is further used to: determine the historical information ranking of the historical recall information on the data node according to the distribution of the data node; and average or weight the historical information ranking to obtain the node ranking of the data node.
[0092] In one embodiment of the present disclosure, the acquisition module 402 is further used to: call a pre-trained ranking prediction model, and input the distribution of the data nodes into the ranking prediction model, and the ranking prediction model outputs the node rankings of different data nodes.
[0093] In one embodiment of the present disclosure, the optimization module 403 is also used to: determine at least one target data node to which the candidate information belongs, and the initial first information ranking of the candidate information on the target data node; optimize the first information ranking according to the node ranking of the target data node to obtain the second information ranking of the candidate information on the target data node; obtain the global optimized ranking of the candidate information according to the second information ranking of the candidate information on the target data node.
[0094] In one embodiment of the present disclosure, the optimization module 403 is also used to: determine the ranking optimization parameters corresponding to the target data node according to the node ranking of the target data node; optimize the first information ranking according to the ranking optimization parameters to obtain the second information ranking of the candidate information on the target data node.
[0095] In one embodiment of the present disclosure, the optimization module 403 is further used to: map the node ranking of the target data node to determine a mapping value within a target numerical range; and determine the mapping value as a ranking optimization parameter corresponding to the target data node.
[0096] In an embodiment of the present application, the distribution of historical recall information on data nodes can be used to predict the ranking of data nodes. Furthermore, based on the ranking of data nodes, the ranking of searched information can be optimized or adjusted, thereby using the ranking of data nodes as a reference factor for the global rank. This can increase global reference factors in local multi-way searches, thereby achieving the goal of balancing memory and network consumption. That is, the accuracy of search results can be enhanced without increasing memory and network consumption.
[0097] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0098] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
[0099] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.
[0100] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program / instruction stored in a read-only memory (ROM) 502 or a computer program / instruction loaded from a storage unit 506 to a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.
[0101] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506 such as a keyboard, a mouse, etc.; an output unit 507 such as various types of displays, speakers, etc.; a storage unit 508 such as a disk, an optical disk, etc.; and a communication unit 509 such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0102] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as information retrieval methods for database scenarios. For example, in some embodiments, the information retrieval method for database scenarios may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 506. In some embodiments, part or all of the computer program / instructions may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program / instructions are loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the information retrieval method for the database scenario described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to execute the information retrieval method for a database scenario in any other appropriate manner (for example, by means of firmware).
[0103] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs / instructions that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0104] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
[0105] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0106] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).
[0107] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.
[0108] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs / instructions running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.
[0109] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in the disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in the disclosure can be achieved, and this document does not limit them here.
[0110] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.
Claims
1. An information retrieval method for database scenarios, wherein: The method comprises: receiving a recall request, and performing information recall at different data nodes according to the recall request to obtain at least one candidate information to be recalled; Get the node rankings of different data nodes; The candidate information is ranked and optimized based on the node ranking, and target information is fed back to the client according to the optimized ranking of the candidate information.
2. The method according to claim 1, wherein: The obtaining of node rankings of different data nodes includes: Get the historical recall information corresponding to the historical recall request; The node rankings of the different data nodes are determined according to the historical recall information.
3. The method according to claim 2, wherein: Determining the node rankings of the different data nodes according to the historical recall information includes: Determine the distribution of the historical recall information on the different data nodes; The data nodes are ranked according to the distribution of the data nodes to obtain node rankings of the different data nodes.
4. The method according to claim 3, wherein: The ranking of the data nodes according to the distribution of the data nodes to obtain the node rankings of the different data nodes includes: Determine a first information total amount of the historical recall information; Determining the total amount of second information recalled for the data node according to the distribution information of the data node; The data nodes are sorted according to the first total amount of information and the second total amount of information to obtain node rankings of the different data nodes.
5. The method according to claim 4, wherein: The step of sorting the data nodes according to the first total amount of information and the second total amount of information to obtain node rankings of the different data nodes includes: Determining a proportion of the recalled information corresponding to the data node according to the first total amount of information and the second total amount of information; The data nodes are sorted according to the proportion of the recalled information to obtain the node rankings of the different data nodes.
6. The method according to claim 3, wherein: The ranking of the data nodes according to the distribution of the data nodes to obtain the node rankings of the different data nodes includes: Determine the historical information ranking of the historical recall information on the data node according to the distribution of the data node; The historical information rankings are averaged or weighted to obtain the node rankings of the data nodes.
7. The method according to claim 3, wherein: The ranking of the data nodes according to the distribution of the data nodes to obtain the node rankings of the different data nodes includes: A pre-trained ranking prediction model is called, and the distribution of the data nodes is input into the ranking prediction model, and the ranking prediction model outputs the node rankings of different data nodes.
8. The method according to any one of claims 1 to 7, wherein: The optimizing the ranking of the candidate information based on the node ranking includes: Determine at least one target data node to which the candidate information belongs, and an initial first information ranking of the candidate information on the target data node; Optimizing the first information ranking according to the node ranking of the target data node to obtain a second information ranking of the candidate information on the target data node; According to the second information ranking of the candidate information on the target data node, a global optimized ranking of the candidate information is obtained.
9. The method according to claim 8, wherein: The step of optimizing the first information ranking according to the node ranking of the target data node to obtain the second information ranking of the candidate information on the target data node includes: Determining a ranking optimization parameter corresponding to the target data node according to the node ranking of the target data node; The first information ranking is optimized according to the ranking optimization parameter to obtain a second information ranking of the candidate information on the target data node.
10. The method according to claim 9, wherein: The step of determining the ranking optimization parameter corresponding to the target data node according to the node ranking of the target data node includes: Mapping the node ranking of the target data node to determine a mapping value within a target numerical range; Determine that the mapping value is a ranking optimization parameter corresponding to the target data node.
11. An information retrieval device for database scenarios, wherein: The device comprises: A recall module, used to receive a recall request, and perform information recall at different data nodes according to the recall request to obtain at least one candidate information to be recalled; The acquisition module is used to obtain the node rankings of different data nodes; The optimization module is used to optimize the ranking of the candidate information based on the node ranking, and feed back target information to the client according to the optimized ranking of the candidate information.
12. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.
13. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-10.
14. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 10 is implemented.