Generalized dictionary generation methods, devices, electronic equipment, and storage media

Abstract

This disclosure provides a method, apparatus, electronic device, storage medium, and program product for generating a generalized dictionary, relating to the field of computer technology, and particularly to the fields of natural language processing (NLP), big data, and intelligent search technology. The specific implementation scheme is as follows: Search terms and corresponding ad-triggered words from search results are obtained from historical data, resulting in a set of text pairs composed of the search terms and ad-triggered words; in the set of text pairs, ad-triggered words matching the search terms are extracted as the generalized dictionary of the search terms based on the text matching relationship between the search terms and ad-triggered words. This disclosure can improve the relevance between generalized words and search terms in the generalized dictionary, and improve the accuracy of the generalized words.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more particularly to the fields of natural language processing (NLP), big data, and intelligent search technology, specifically to a generalized dictionary generation method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] In the process of recalling ads based on users' search terms, it is necessary to use a generalized dictionary of search terms to obtain more accurate and richer ad recall results.

[0003] Current technologies typically use core word clustering to obtain a generalized dictionary. However, user needs are usually quite broad and lack a specific core semantic meaning. Therefore, the semantic relevance of the generalized words differs significantly from the original search terms, affecting the accuracy of the generalized words. Summary of the Invention

[0004] This disclosure provides a generalized dictionary generation method, apparatus, electronic device, storage medium, and program product.

[0005] According to one aspect of this disclosure, a method for generating a generalized dictionary is provided, comprising:

[0006] Extract search terms and corresponding ad delivery terms triggered by users from historical data to obtain a set of text pairs consisting of the search terms and ad delivery terms.

[0007] In the set of text pairs, based on the text matching relationship between the search term and the target term, the target term that matches the search term is extracted as a generalized dictionary of the search term.

[0008] According to another aspect of this disclosure, a generalized dictionary generation apparatus is provided, comprising:

[0009] The text pair set acquisition module is used to obtain search terms and corresponding ad delivery terms triggered by users in the search results from historical data, and obtain a text pair set composed of the search terms and ad delivery terms;

[0010] The generalization dictionary acquisition module is used to extract, from the text pair set, the target words that match the search term as the generalization dictionary of the search term, based on the text matching relationship between the search term and the target word.

[0011] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0012] At least one processor; and

[0013] A memory communicatively connected to the at least one processor; wherein,

[0014] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the generalized dictionary generation method described in any embodiment of this disclosure.

[0015] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform the generalized dictionary generation method described in any embodiment of this disclosure.

[0016] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the generalized dictionary generation method described in any embodiment of this disclosure.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0018] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0019] Figure 1 This is a schematic diagram of a generalized dictionary generation method according to an embodiment of the present disclosure;

[0020] Figure 2 This is a schematic diagram of a generalized dictionary generation method according to an embodiment of the present disclosure;

[0021] Figure 3 This is a schematic diagram of a generalized dictionary generation method according to an embodiment of the present disclosure;

[0022] Figure 4 This is a schematic diagram of a generalized dictionary generation method according to an embodiment of the present disclosure;

[0023] Figure 5 This is a schematic diagram of a generalized dictionary generation apparatus according to an embodiment of the present disclosure;

[0024] Figure 6 This is a block diagram of an electronic device used to implement the generalized dictionary generation method of the embodiments of this disclosure. Detailed Implementation

[0025] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0026] Figure 1 This is a flowchart illustrating a generalized dictionary generation method according to an embodiment of the present disclosure. This embodiment is applicable to situations where user-input search terms are generalized, and relates to the field of computer technology, particularly natural language processing (NLP), big data, and intelligent search technology. The method can be executed by a generalized dictionary generation device, which is implemented in software and / or hardware, preferably configured in an electronic device, such as a computer or server. Figure 1 As shown, the method specifically includes the following:

[0027] S101. Obtain the search terms and the corresponding ad delivery terms triggered by users in the search results from historical data to obtain a set of text pairs consisting of search terms and ad delivery terms.

[0028] S102. In the text pair set, based on the text matching relationship between the search term and the target term, extract the target term that matches the search term as a generalization dictionary of the search term.

[0029] Historical data refers to the user's search history, including the search terms entered by the user and the keywords used to deliver advertisements triggered by the user in the search results. For example, advertisers purchase keywords corresponding to their advertisements in the advertising system. When a user's search term matches a keyword, the advertising system recalls and delivers the advertisement. If the user sees the advertisement in the search results and clicks on it, it means that the user has triggered the advertisement.

[0030] It's important to note that to improve the accuracy of ad recall, a pre-built generalization dictionary is typically used to generalize the user-input search terms during the ad recall process. It's crucial to ensure that the generalized terms are semantically relevant to the original search terms. Then, relevant algorithms are used to recall ads based on the generalized search terms. Current technologies often use core word clustering to obtain the generalization dictionary for search terms. However, user-input search terms usually lack clear core semantics, making it difficult to generate accurate generalized terms. This results in significant differences in semantic relevance between the generalized terms and the original search terms, affecting the accuracy of the generalized terms.

[0031] In the technical solution of this embodiment, firstly, search terms and corresponding ad delivery terms triggered by users in the search results are obtained from historical data, resulting in a text pair set composed of search terms and ad delivery terms. In the text pair set, the same search term can correspond to multiple different ad delivery terms, and the same ad delivery term can also correspond to multiple different search terms, each forming a different text pair. Then, in the text pair set, based on the text matching relationship between search terms and ad delivery terms, ad delivery terms that match the search terms are extracted as a generalized dictionary of search terms.

[0032] In this embodiment, since the target words in the text pair set are the target words corresponding to the ads clicked by users, each text pair consisting of the search term and the target word represents the original data of user searches and clicks in real historical situations. Furthermore, in each text pair set, the search term and the target word have a certain semantic relevance. Considering that ad recall is based on a generalized dictionary, if the generalized words themselves originate from the target words of the ads, the quality of the final ad recall can be improved. Therefore, in this embodiment, based on the text matching relationship between the search term and the target word, the target words matching the search term are extracted from the text pair set as the generalized dictionary of the search term. This ensures that the obtained generalized words have a certain semantic relevance to the search term and improves the accuracy of the generalized dictionary, thereby enhancing the effect of subsequent ad recall and increasing ad click-through rate. Text matching can be achieved using existing natural language processing techniques. Regarding how to determine a match with the search term, it can be completely identical, partially identical, or semantically identical or similar; this embodiment does not impose any limitations on this, and corresponding standards and rules can be configured according to actual needs.

[0033] In one implementation, within the text pair set, based on the text matching relationship between the search term and the target term, target terms that match the search term are extracted as a generalization dictionary for the search term. This can include: matching the search term in the text pair set with the target term, extracting a first set of target terms containing the search term based on the matching results, and using the first set of target terms as the generalization dictionary for the search term. Specifically, within the text pair set, each search term is matched with all target terms. If any target term contains the current search term in its text, that target term is used as a generalization term for the current search term, and the generalization term corresponding to each search term is finally obtained, thereby generating a generalization dictionary. Since the generalization term is a target term that contains the search term, the semantic relevance between the generalization term and the search term is ensured.

[0034] Figure 2 This is a flowchart illustrating a generalized dictionary generation method according to an embodiment of the present disclosure. This embodiment further optimizes the above embodiments. Figure 2 As shown, the method specifically includes the following:

[0035] S201. Obtain the search terms and the corresponding ad delivery terms triggered by users in the search results from historical data to obtain a set of text pairs consisting of search terms and ad delivery terms.

[0036] S202. Remove regional terms and special characters from search terms and target terms.

[0037] Advertisers often pad their ad keywords with ineffective terms, such as adding place names or special characters. Place names include city or town names, while special characters include underscores or dashes. However, this results in a limited generalization pool due to an excessive number of similar and ineffective keywords. Therefore, in this embodiment, place names and special characters are removed from ad keywords before obtaining the generalization dictionary, and special characters can be replaced with empty strings. Similarly, user-input search terms may also contain place names or special characters, which also affects the generalization pool. Therefore, it is necessary to remove place names and special characters from search terms to obtain "clean" search terms.

[0038] S203. Extract the core words of the search terms from the text pair set.

[0039] In this process, any existing natural language processing technique can be used to extract core words, and this disclosure does not impose any restrictions on it.

[0040] S204. Perform text matching between the core keywords and the target keywords, and extract a second set of target keywords containing the core keywords based on the matching results.

[0041] S205. Use the second set of target terms as a generalized dictionary of search terms.

[0042] Specifically, the technical solution of this disclosure uses core word matching to extract target words containing the core words of the search term from a set of text pairs to form a second target word set. Since user-input search terms are often arbitrary, they can easily contain words irrelevant to the search purpose, affecting the accuracy of ad recall. Therefore, extracting the core words of the search term and using core word matching to obtain the second target word set makes the generalized data structure more accurate and richer, further improving the quantity and quality of generalized words.

[0043] In addition, in one implementation, the first set of target words and the second set of target words obtained in the above embodiments can be merged to generate a generalized dictionary, thereby further increasing the number of generalized dictionaries.

[0044] The technical solution of this disclosure removes "unclean" parts such as place names and special characters from search terms and target terms through data processing to obtain accurate search terms and target terms. Then, it performs generalization processing on the set based on text and increases the number of generalized terms through core word matching, while ensuring the relevance between generalized terms and search terms.

[0045] Figure 3 This is a flowchart illustrating a generalized dictionary generation method according to an embodiment of the present disclosure. This embodiment further optimizes the above embodiments. Figure 3 As shown, the method specifically includes the following:

[0046] S301. Obtain the search terms and the corresponding ad delivery terms triggered by users in the search results from historical data to obtain a set of text pairs consisting of search terms and ad delivery terms.

[0047] S302. Remove regional terms and special characters from search terms and target terms.

[0048] S303. In the text pair set, perform text matching between the search term and the target term, and extract the third target term set containing the search term based on the matching results.

[0049] S304. Perform text matching between the delivery words in the third delivery word set and the search words in the text pair set, and extract the first search word set containing the delivery words in the third delivery word set based on the matching results.

[0050] S305. Perform text matching between the search terms in the first search term set and the delivery terms in the text pair set, and extract the fourth delivery term set containing the search terms in the first search term set based on the matching results.

[0051] S306. Use the fourth set of delivery terms as a generalized dictionary of the search terms in the text pair set.

[0052] Specifically, in this embodiment, a deeper level of generalized terms is constructed according to the generalization path of "search term - target term - search term - target term". Similarly, text matching is performed within the text pair set. First, a third target term set containing the search term is extracted. Then, each target term in the third target term set is matched against the text pair set to extract a first search term set containing the target terms in the third target term set. Next, each search term in the first search term set is matched against the text pair set to extract a fourth target term set containing the target terms in the first search term set. This fourth target term set can then serve as a deeper level of generalized terms for the initial search term, thereby increasing the number of generalized terms in the generalization dictionary. Furthermore, since the generalized terms are obtained through text matching within the text pair set according to the aforementioned generalization path, it ensures that the generalized terms have a certain semantic relevance to the corresponding search term.

[0053] In addition, in one implementation, the first set of target words, the second set of target words, and the third set of target words obtained in the above embodiments can be merged to generate a generalized dictionary, thereby further increasing the number of generalized dictionaries. Figure 4 This is a schematic diagram of a generalized dictionary generation method according to an embodiment of the present disclosure, such as... Figure 4 As shown, binary co-occurrence path generalization refers to extracting the target words containing the search term from the obtained text pair set as generalized words for the search term; core word ternary path generalization refers to extracting the target words containing the core word of the search term as generalized words for the search term; and quaternary path generalization refers to constructing deeper generalized words according to the generalization path of "search term-target word-search term-target word". Through the above different co-occurrence generalization strategies, more different generalized words can be obtained, enriching the generalization dictionary while ensuring that the generalized words have a certain semantic relevance to the search term.

[0054] Furthermore, the method in this embodiment of the generalization dictionary may further include: determining the value score of each generalization word based on its historical click count, impression count, and advertiser bidding information; and sorting the generalization words according to their value scores. Specifically, to obtain the advertising value of words with strong user click-through rates and monetization capabilities, the click and impression counts of ads under each generalization word can be statistically analyzed, and all historical bidding information corresponding to each generalization word can be calculated. Finally, the average impression price under each generalization word is calculated as the standard for offline value ranking of all generalization words, thereby obtaining the order of each generalization word. It should be noted that this embodiment of the disclosure does not limit the specific algorithm for determining the value score and can be configured as needed.

[0055] The technical solution of this disclosure generalizes search terms based on the post-click co-occurrence data of user-input search terms and advertiser-placed ad terms, thus obtaining a generalized dictionary of search terms. This addresses the problem of poor ad performance in existing technologies due to insufficient relevance between generalized terms and search terms, by leveraging user click behavior in real-world environments. Furthermore, it allows for convenient adjustment of the generalized dictionary's ranking based on user post-click data, further enhancing traffic monetization capabilities and increasing platform advertising revenue.

[0056] Figure 5 This is a schematic diagram of a generalized dictionary generation apparatus according to an embodiment of this disclosure. This embodiment is applicable to situations where user-input search terms are generalized, and relates to the field of computer technology, particularly to the fields of natural language processing (NLP), big data, and intelligent search technology. This apparatus can implement the generalized dictionary generation method described in any embodiment of this disclosure. Figure 5 As shown, the device 500 specifically includes:

[0057] The text pair set acquisition module 501 is used to acquire search terms and corresponding user-triggered advertising words from historical data, and obtain a text pair set composed of the search terms and advertising words;

[0058] The generalization dictionary acquisition module 502 is used to extract, from the text pair set, the target words that match the search term as the generalization dictionary of the search term, based on the text matching relationship between the search term and the target word.

[0059] Optionally, the device further includes:

[0060] The data processing module is used to remove regional terms and special characters from the search terms and the target terms before the generalization dictionary acquisition module 502 extracts the target terms that match the search terms as the generalization dictionary of the search terms.

[0061] Optionally, the generalized dictionary acquisition module 502 includes a first generalization submodule, which includes:

[0062] The first target word set extraction unit is used to perform text matching between the search term and the target word in the text pair set, and extract the first target word set containing the search term based on the matching result;

[0063] The first generalization dictionary acquisition unit is used to use the first set of target words as the generalization dictionary of the search terms.

[0064] Optionally, the generalized dictionary acquisition module 502 includes a second generalization submodule, which includes:

[0065] The core word extraction unit is used to extract the core words of the search terms in the text pair set;

[0066] The second target word set extraction unit is used to perform text matching between the core word and the target word, and extract a second target word set containing the core word based on the matching result;

[0067] The second generalization dictionary acquisition unit is used to use the second set of target words as the generalization dictionary for the search terms.

[0068] Optionally, the generalized dictionary acquisition module 502 includes a third generalization submodule, which includes:

[0069] The third target word set extraction unit is used to perform text matching between the search term and the target word in the text pair set, and extract the third target word set containing the search term based on the matching result;

[0070] The first search term set extraction unit is used to perform text matching between the delivery terms in the third delivery term set and the search terms in the text pair set, and extract the first search term set containing the delivery terms in the third delivery term set based on the matching results.

[0071] The fourth delivery word set extraction unit is used to perform text matching between the search words in the first search word set and the delivery words in the text pair set, and extract the fourth delivery word set containing the search words in the first search word set based on the matching results;

[0072] The third generalization dictionary acquisition unit is used to use the fourth set of delivery words as a generalization dictionary for the search terms in the text pair set.

[0073] Optionally, the device further includes a sorting module for:

[0074] The value score of each generalized word is determined based on its historical click count, impression count, and advertiser bidding information in the generalized dictionary.

[0075] The generalized words are sorted according to their value scores.

[0076] The above-described products can perform the methods provided in any embodiment of this disclosure, and have the corresponding functional modules and beneficial effects for performing the methods.

[0077] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in this technical solution comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0078] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0079] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments 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 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0080] like Figure 6 As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 602 or a computer program loaded from storage unit 608 into random access memory (RAM) 603. RAM 603 may also store various programs and data required for the operation of device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0081] Multiple components in device 600 are connected to I / O interface 605, including: input unit 606, such as keyboard, mouse, etc.; output unit 607, such as various types of monitors, speakers, etc.; storage unit 608, such as disk, optical disk, etc.; and communication unit 609, such as network card, modem, wireless transceiver, etc. Communication unit 609 allows device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0082] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as the generalized dictionary generation method. For example, in some embodiments, the generalized dictionary generation method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the generalized dictionary generation method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the generalized dictionary generation method by any other suitable means (e.g., by means of firmware).

[0083] Various embodiments 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-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0084] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0085] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0086] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0087] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0088] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is established by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem that addresses the management difficulties and weak business scalability inherent in traditional physical hosting and VPS services. Servers can also be servers for distributed systems or servers integrated with blockchain technology.

[0089] Artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, and big data processing. AI software technologies mainly include computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graph technologies.

[0090] Cloud computing refers to a technology system that enables access to a shared pool of physical or virtual resources via a network. These resources can include servers, operating systems, networks, software, applications, and storage devices, and can be deployed and managed on demand and in a self-service manner. Cloud computing technology can provide efficient and powerful data processing capabilities for applications such as artificial intelligence and blockchain, as well as for model training.

[0091] 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 described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution provided in this disclosure can be achieved, and this is not limited herein.

[0092] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for generating a generalized dictionary, comprising: Extract search terms and corresponding ad delivery terms triggered by users from historical data to obtain a set of text pairs consisting of the search terms and ad delivery terms. In the set of text pairs, based on the text matching relationship between the search term and the target term, the target terms that match the search term are extracted as a generalized dictionary of the search term, including: In the set of text pairs, the search term and the target term are matched, and a third set of target terms containing the search term is extracted based on the matching results; The terms in the third set of terms are matched with the search terms in the text pair set, and a first set of search terms containing the terms in the third set of terms is extracted based on the matching results. The search terms in the first search term set are matched with the delivery terms in the text pair set, and a fourth delivery term set containing the search terms in the first search term set is extracted based on the matching results. The fourth set of delivery terms is used as a generalized dictionary of the search terms in the text pair set.

2. The method according to claim 1, before extracting the target words that match the search term as a generalized dictionary of the search term, the method further includes: Remove regional terms and special characters from the search terms and target terms.

3. The method according to claim 1, wherein, The step of extracting, based on the text matching relationship between the search term and the target term, target terms that match the search term in the text pair set, as a generalization dictionary for the search term, further includes: In the set of text pairs, the search term and the target term are matched, and a first set of target terms containing the search term is extracted based on the matching result; The first set of target terms is used as the generalized dictionary of the search terms.

4. The method according to claim 1, wherein, The step of extracting, based on the text matching relationship between the search term and the target term, target terms that match the search term in the text pair set, as a generalization dictionary for the search term, further includes: Extract the core words of the search terms from the text pair set; The core words are matched with the target words, and a second set of target words containing the core words is extracted based on the matching results. The second set of target terms is used as the generalized dictionary of the search terms.

5. The method according to claim 1, further comprising: The value score of each generalized word is determined based on its historical click count, impression count, and advertiser bidding information in the generalized dictionary. The generalized words are sorted according to their value scores.

6. A generalized dictionary generation apparatus, comprising: The text pair set acquisition module is used to obtain search terms and corresponding ad delivery terms triggered by users in the search results from historical data, and obtain a text pair set composed of the search terms and ad delivery terms; The generalization dictionary acquisition module is used to extract, from the text pair set, the target words that match the search term as the generalization dictionary of the search term, based on the text matching relationship between the search term and the target word; The generalized dictionary acquisition module includes: a third generalization submodule; The third generalization submodule includes: The third target word set extraction unit is used to perform text matching between the search term and the target word in the text pair set, and extract the third target word set containing the search term based on the matching result; The first search term set extraction unit is used to perform text matching between the delivery terms in the third delivery term set and the search terms in the text pair set, and extract the first search term set containing the delivery terms in the third delivery term set based on the matching results. The fourth delivery word set extraction unit is used to perform text matching between the search words in the first search word set and the delivery words in the text pair set, and extract the fourth delivery word set containing the search words in the first search word set based on the matching results; The third generalization dictionary acquisition unit is used to use the fourth set of delivery words as a generalization dictionary for the search terms in the text pair set.

7. 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 to enable the at least one processor to perform the generalized dictionary generation method according to any one of claims 1-5.

8. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the generalized dictionary generation method according to any one of claims 1-5.

9. A computer program product comprising a computer program / instructions that, when executed by a processor, implement the generalized dictionary generation method according to any one of claims 1-5.

Images