Business intent retrieval method, system, device and computer readable storage medium

Abstract

The application provides a service intention retrieval method, system, device and computer readable storage medium, the method comprising: receiving a retrieval appeal; if the received retrieval appeal can be split, splitting the retrieval appeal into several words, and identifying the parts of speech of each word; setting a part of speech priority for each word, and matching a retrieval weight corresponding to the priority for different part of speech priorities; detecting the words with active tags or passive tags in the retrieval appeal after setting the part of speech priority, if there are words with active tags, then increasing the retrieval weight of the retrieval sentence containing the active tags; if there are words with passive tags, then increasing the retrieval weight of the retrieval sentence containing the passive tags; calling a search engine to retrieve the retrieval sentence to obtain a retrieval intention matching the retrieval appeal. The application analyzes the real retrieval intention of the user, improves the accuracy of user retrieval, the search usage rate and the rationality of the search result.

Description

Technical Field

[0001] The present invention belongs to the technical field of Internet search, and relates to a method and a system, in particular to a method, a system, a device and a computer-readable storage medium for retrieving business intentions. Background Art

[0002] With the booming development of the Internet industry, precise data marketing can provide strong support for the development of enterprises. Under the concept of data marketing, the search module has become an effective means to improve the user experience and direct conversion. However, in the industry, the search in identifying user intentions is at a relatively primary stage. An excellent search intention recognition algorithm for insurance business can boost the sales of insurance products and enhance the retention of users on the app.

[0003] The current deficiencies in search intention recognition in the industry are as follows:

[0004] 1. It is unable to identify intentions for specific references, for example, in the insurance industry:

[0005] For example, when a user searches for "e-life insurance", a traditional search engine will display data containing the three characters "e", "sheng", and "bao", and the data is very messy. Even results containing only one of these characters may be retrieved.

[0006] 2. It is unable to perform precise retrieval according to specific requirements of users:

[0007] For example, when a user searches for "insurance suitable for children", a traditional search engine will only recognize "insurance" and cannot identify the retrieval requirement of the user for a specific group of people, namely "children".

[0008] The above are common problems in search intention recognition in the insurance industry. If not solved, it will directly affect the user experience. It should be noted that users' search behaviors generally have precise purposes and they want to quickly find products and information that meet their personal demands through search. A search engine that cannot accurately identify intentions will be a failed product.

[0009] Therefore, how to provide a method, a system, a device and a computer-readable storage medium for retrieving business intentions to solve the defects that the prior art cannot meet specific requirements of users and the retrieval of products is inaccurate has actually become a technical problem urgently to be solved by those skilled in the art. Summary of the Invention

[0010] In view of the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide a method, a system, a device and a computer-readable storage medium for retrieving business intentions to solve the problem that the prior art cannot perform precise retrieval according to specific requirements of users.

[0011] To achieve the above and other related objectives, in a first aspect, the present invention provides a method for retrieving business intent, comprising:

[0012] Receive a search request to be retrieved; the search request contains terms used to search for business intent;

[0013] If the received search request can be broken down, the search request is broken down into several words, and the part of speech of each word is collected and identified.

[0014] Assign part-of-speech priority to each word and match search weights that match the priority to different part-of-speech priorities;

[0015] The system detects whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, the system searches for search statements containing active tags and increases the search weight of search statements containing active tags. If there are words with passive tags, the system searches for search statements containing passive tags and increases the search weight of search statements containing passive tags. Specifically, active tags are configured for passively received words, and passive tags are configured for actively identified words.

[0016] Search queries are displayed in priority according to their increased search weight, and the search engine is invoked to retrieve the search intent that matches the search request.

[0017] Secondly, the present invention provides a business intent retrieval system, comprising:

[0018] A receiving module is used to receive a search request to be retrieved; the search request includes words used to express the search intent.

[0019] The language processing module is used to, if the received search request can be split, split the search request into several words, collect them and identify the part of speech of each word;

[0020] The priority setting module is used to set the part-of-speech priority for each word and match the search weight that matches the priority for different parts of speech.

[0021] The retrieval execution module is used to detect whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, the module retrieves search statements containing active tags and increases their search weight. If there are words with passive tags, the module retrieves search statements containing passive tags and increases their search weight. The module prioritizes and displays the corresponding search statements according to their increased search weights, and calls the search engine to retrieve the search statements to obtain search intents that match the search request.

[0022] Among them, active labels are configured for passively received words, and passive labels are configured for actively identified words.

[0023] Thirdly, the present invention provides a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the retrieval method for the business intent.

[0024] Fourthly, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the retrieval method for the business intent.

[0025] As described above, the service identification method / system, service retrieval method / system, medium, and device of the present invention have the following beneficial effects:

[0026] The business intent retrieval method, system, device, and computer-readable storage medium described in this invention parses the user's true search intent, providing basic tags and part-of-speech classifications for the next step of search engine retrieval, thus helping to improve the accuracy of user searches. This invention can obtain the core keywords of user searches and match them with relevant product tags, displaying related products and information, satisfying effective user searches in various business fields, and significantly improving search usage and the rationality of search results. Attached Figure Description

[0027] Figure 1A The diagram shown is a flowchart illustrating a specific implementation of the business intent retrieval method of the present invention in one embodiment.

[0028] Figure 1B The diagram shown is a flowchart of S17 in the business intent retrieval method of the present invention.

[0029] Figure 2 The diagram shown is a schematic representation of the principle structure of the business intent retrieval system of the present invention in one embodiment.

[0030] Figure 3 The diagram shown is a structural schematic of a computer device according to an embodiment of the present invention.

[0031] Component designation explanation

[0032] Detailed Implementation

[0033] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0034] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0035] Example 1

[0036] This embodiment provides a method for retrieving business intent, including:

[0037] Receive a search request; the search request contains terms used to search for business intent;

[0038] If the received search request can be broken down, the search request is broken down into several words, and the part of speech of each word is collected and identified.

[0039] Assign part-of-speech priority to each word and match search weights that match the priority to different part-of-speech priorities;

[0040] The system detects whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, the system searches for search statements containing active tags and increases the search weight of search statements containing active tags. If there are words with passive tags, the system searches for search statements containing passive tags and increases the search weight of search statements containing passive tags. Specifically, active tags are configured for passively received words, and passive tags are configured for actively identified words.

[0041] Search queries are displayed in priority according to their increased search weight, and the search engine is invoked to retrieve the search intent that matches the search request.

[0042] The following will describe in detail the business intent retrieval method provided in this embodiment with reference to the illustrations. This business intent retrieval method can be applied to various target industry sectors, accurately retrieving the user's true business intent by parsing the user's input search request. The following will describe in detail the application of the business identification method described in this embodiment to the insurance industry sector.

[0043] Please see Figure 1A The diagram shows a specific implementation flow of a method for retrieving business intent in one embodiment. For example... Figure 1A As shown, the service identification method specifically includes the following steps:

[0044] S11, Receive a search request. In this embodiment, the search request includes terms used to search for business intent.

[0045] For example, the search request received was "insurance suitable for children".

[0046] S12, detect whether the received search request can be divided. If yes, execute S13; if no, execute S14, that is, collect indivisible search requests.

[0047] S13, break down the received search request into several words and collect the broken-down words.

[0048] In this embodiment, step S12 uses a trained natural language processing component related to an industry domain to perform segmentation detection and segmentation processing on the received language data.

[0049] Natural Language Processing (NLP) components, such as "HanLP," are Java toolkits comprised of a series of models and algorithms designed to facilitate the application of NLP in production environments. HanLP supports Chinese word segmentation (N-shortest path segmentation, CRF segmentation, indexed segmentation, user-defined dictionaries, part-of-speech tagging), named entity recognition (person names, transliterated person names, place names, entity name recognition), keyword extraction, automatic summarization, phrase extraction, pinyin conversion, simplified / traditional Chinese conversion, text recommendation, and dependency parsing (MaxEnt dependency parsing, neural network dependency parsing), among others.

[0050] For example, when the input language data is "Insurance suitable for children to purchase", the natural language processing component "HanLP" splits "Insurance suitable for children to purchase" into five splittable words: suitable, children, purchase, 's, insurance. S14, if the received language data is not splittable, collect the non - splittable words. For example, when the natural language processing component HanLP receives non - splittable language data such as iKangbao Million Medical Chronic Disease Edition, iKangbao.Million Medical (Chronic Disease Edition), Chronic Disease Edition, Million Medical, Elderly Edition, Upgrade Edition, etc., these language data are collected as non - splittable words.

[0051] For example, after receiving the retrieval request "e - Shengbao" -> "e", "Sheng", "bao", the insurance - specific word "e - Shengbao" will not be split. At this time, the corpus retrieved by the search engine will be the reasonable word segmentation after intention recognition. The search results will only be related to "e - Shengbao" and will not retrieve any invalid entries containing "e", "Sheng", "bao".

[0052] In this embodiment, the non - splittable language data can be received passively (i.e., manually input by business personnel in related industry fields) or actively identified by the business recognition system.

[0053] S15, classify and store the collected words (i.e., splittable words and non - splittable words) as splittable words related to the target industry field and non - splittable words related to the target industry field, and identify the word natures of the splittable words and non - splittable words. In this embodiment, the identified word natures include verbs, nouns, gerunds, numerals, function words, etc.

[0054] For example, split "Insurance suitable for children to purchase" into five splittable words: suitable, children, purchase, 's, insurance and classify them into splittable words related to the insurance industry field.

[0055] For example, classify non - splittable words such as iKangbao Million Medical Chronic Disease Edition, iKangbao Million Medical (Chronic Disease Edition), Chronic Disease Edition, Million Medical, Elderly Edition, Upgrade Edition, etc. into non - splittable words related to the insurance industry field.

[0056] For example, identify the word natures of the five split words suitable, children, purchase, 's, insurance in sequence. That is, the word nature of "suitable" is a verb, the word nature of "children" is a noun, the word nature of "purchase" is a verb, the word nature of "'s" is a particle, and the word nature of "insurance" is a noun.

[0057] For example, identify the non - splittable word "iKangbao Million Medical Chronic Disease Edition" as a proper noun in nouns.

[0058] S16, filter out the split function words (function words include adverbs, prepositions, conjunctions, particles, etc.).

[0059] For example, filter out the particle "de" split from "insurance suitable for children to purchase".

[0060] S17. Set a词性 priority for each word and match a retrieval weight that matches its priority for different词性 priorities.

[0061] Specifically, the S17 includes setting词性 priorities for both splittable words and unsplittable words.

[0062] In this embodiment, to better identify the user's retrieval intention, please refer to Figure 1B , which shows a schematic flow diagram of S17. As Figure 1B shown, the S17 specifically includes:

[0063] S171. Set nouns and nominal verbs as first-priority words, set verbs and personal names, geographical-related words, institutional groups, and proper nouns in nouns as second-priority words, set adjectives as third-priority words, and set numerals as fourth-priority words.

[0064] S172. Match the highest retrieval weight for first-priority words, match the second-highest retrieval weight for second-priority words, match a lower retrieval weight for third-priority words, and match the lowest retrieval weight for fourth-priority words.

[0065] For example, set the noun "life insurance" related to insurance business as a first-level word, the nominal verb "claim settlement" as a first-level word, the abbreviation "三高" as a first-priority word, the name representing time "elderly" as a first-priority word, personal names / place names / institutional groups as second-priority words, and the proper noun "cleansing face" as a second-priority word.

[0066] For example, the S17 sets "suitable" split from the split words of "insurance suitable for children to purchase" as a second-priority word, "children" as a first-priority word, "purchase" as a second-priority word, "insurance" as a first-priority word, and configures the highest retrieval weight for "children" and "insurance", for example, configure 1.5 as the highest retrieval weight, and configure the second-highest retrieval weight for "suitable" and "purchase", for example, configure 1.2 as the second-highest retrieval weight.

[0067] It should be noted that the term "词性" in the original text seems to be an incorrect or incomplete expression. It might be a specific term in Chinese that needs to be accurately defined or corrected for a more precise translation. Here, it is tentatively translated as "词性" for the purpose of following the translation rules.S18: Detect whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, search for search statements containing active tags and increase their search weight; if there are words with passive tags, search for search statements containing passive tags and increase their search weight. Specifically, active tags are configured for passively received words, and passive tags are configured for actively identified words.

[0068] In this embodiment, active tags are configured for passively received words, and passive tags are configured for actively identified words; the active tags are manually defined attributes that operators extract and summarize based on the characteristics of the search results. The passive tags are the inherent attributes of the search results themselves.

[0069] For example, keywords with active tags include overseas medical treatment for cancer, hospital appointments, translation, economy class, transportation costs, overseas accommodation costs, and star-rated hotels, while keywords with passive tags include family, children, overseas high-end medical care, transportation, and accommodation. The insurance product that matches these tags is the Cancer Guardian Overseas Edition.

[0070] Specifically, the step of retrieving search statements containing active tags and increasing the search weight of search statements containing active tags in S18 includes: when the active tag is retrieved, the search weight of the matched active tag is weighted; the step of retrieving search statements containing passive tags and increasing the search weight of search statements containing passive tags includes: when the passive tag is retrieved, the search weight of the matched passive tag is weighted.

[0071] In this embodiment, the step of weighting the retrieval weights of the matched active tags includes: multiplying the retrieval weights of the matched active tags by the highest retrieval weight of the first priority word match.

[0072] In this embodiment, the step of weighting the retrieval weights of the matched passive tags includes multiplying the retrieval weights of the matched passive tags by the secondary higher retrieval weights of the second priority words.

[0073] For example, when searching for the active tag "children," the search weight of insurance products matching the "children" tag is multiplied by the highest search weight for the first priority keyword match, for example, 1.5 times. When searching for passive tags, the weight is multiplied by the second highest search weight for the second priority keyword match, for example, 1.2 times. The result is a weighting factor based on the search engine relevance score multiplied by the tag, which can significantly improve search accuracy.

[0074] S19, prioritize displaying the corresponding search statements according to the improved search weight, and call the search engine to search the search statements to obtain the search intent that matches the search request.

[0075] In this embodiment, in step S19, the items are sorted by default in descending order of search weight, with those listed earlier being more relevant to the user's search intent.

[0076] The business intent retrieval method described in this embodiment analyzes the user's true search intent, providing basic tags and part-of-speech classifications for the next step of the search engine retrieval, which helps improve the accuracy of user searches. Furthermore, this invention can obtain the core keywords of the user's search and match them with relevant product tags, displaying related products and information, satisfying effective user searches in various business fields, and significantly improving search usage and the rationality of search results.

[0077] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following... Figure 1A The method for retrieving business intent, when the computer program is executed by the processor, implements the following steps:

[0078] Receive language data to be retrieved; the language data contains vocabulary used to retrieve business intent;

[0079] If the received language data can be split, the language data is split into several words, collected, and the part of speech of each word is identified.

[0080] Assign part-of-speech priority to each word and match search weights that match the priority to different part-of-speech priorities;

[0081] The system detects whether there are words with active or passive tags in the language data after setting part-of-speech priority. If there are words with active tags, the system retrieves search statements containing active tags and increases the search weight of search statements containing active tags. If there are words with passive tags, the system retrieves search statements containing passive tags and increases the search weight of search statements containing passive tags. Specifically, active tags are configured for passively received words, and passive tags are configured for actively identified words.

[0082] Users are prioritized for display based on their increased search weight to understand their search intent.

[0083] At any possible level of technical detail, this application can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this application.

[0084] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, (but not limited to) electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0085] The computer-readable program described herein can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer-readable program instructions from the network and forwards these instructions to a computer-readable storage medium in the respective computing / processing device. The computer program instructions used to perform the operations of this application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and procedural programming languages ​​such as "C" or similar programming languages. Computer-readable program instructions may execute entirely on a user's computer, partially on a user's computer, as a standalone software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of this application.

[0086] Example 2

[0087] This embodiment also provides a business intent retrieval system, including:

[0088] A receiving module is used to receive a search request; the search request includes words used to express the search intent.

[0089] The language processing module is used to, if the received search request can be split, split the search request into several words, collect them and identify the part of speech of each word;

[0090] The priority setting module is used to set the part-of-speech priority for each word and match the search weight that matches the priority for different parts of speech.

[0091] The retrieval execution module is used to detect whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, the module retrieves search statements containing active tags and increases their search weight. If there are words with passive tags, the module retrieves search statements containing passive tags and increases their search weight. The module prioritizes and displays the corresponding search statements according to their increased search weights, and calls the search engine to retrieve the search statements to obtain search intents that match the search request.

[0092] Among them, active labels are configured for passively received words, and passive labels are configured for actively identified words.

[0093] The following will describe in detail the business intent retrieval system provided in this embodiment, with reference to the illustrations. Please refer to... Figure 2 The diagram shows the principle structure of a business intent retrieval system in one embodiment. Figure 2 As shown, the business intent retrieval system 2 includes a receiving module 21, a language processing module 22, a classification storage module 23, a filtering module 24, a priority setting module 25, and a retrieval execution module 26.

[0094] The receiving module 21 is used to receive search requests. In this embodiment, the search request includes terms used to search for business intent.

[0095] The language processing module 22 is used to detect whether the received search request can be divided. If so, the received search request is divided into several words and the divided words are collected; if not, the indivisible search request is collected.

[0096] In one embodiment, the language processing module 22 uses a trained natural language processing component that is relevant to an industry domain to process the received search request.

[0097] Natural Language Processing (NLP) components, such as "HanLP," are Java toolkits comprised of a series of models and algorithms designed to facilitate the application of NLP in production environments. HanLP supports Chinese word segmentation (N-shortest path segmentation, CRF segmentation, indexed segmentation, user-defined dictionaries, part-of-speech tagging), named entity recognition (person names, transliterated person names, place names, entity name recognition), keyword extraction, automatic summarization, phrase extraction, pinyin conversion, simplified / traditional Chinese conversion, text recommendation, and dependency parsing (MaxEnt dependency parsing, neural network dependency parsing), among others.

[0098] In this embodiment, the indivisible search request can be passively received (i.e., manually entered by business personnel in the relevant industry field) or actively identified by the business identification system.

[0099] In one embodiment, the language processing module 22 is further configured to identify the parts of speech of separable and indivisible words. In this embodiment, the identified parts of speech include verbs, nouns, gerunds, numerals, and function words, etc.

[0100] The classification storage module 23 is used to classify and store the collected words (i.e., divisible words and indivisible words) into divisible words and indivisible words related to the target industry field.

[0101] The filtering module 24 is used to filter the split function words (function words include adverbs, prepositions, conjunctions, auxiliary words, etc.).

[0102] The priority configuration module 25 is used to set part-of-speech priority for each word and match retrieval weights that match different part-of-speech priorities.

[0103] Specifically, the priority configuration module 25 is used to set part-of-speech priorities for both separable and indivisible words.

[0104] In one embodiment, to better identify the user's search intent, the priority configuration module 25 is used to set nouns and noun verbs as first priority words, verbs and nouns including personal names, geographically related words, organizational and group words and proper nouns as second priority words, adjectives as third priority words, and numerals as fourth priority words; the highest search weight is matched for the first priority words, the second highest search weight is matched for the second priority words, the lower search weight is matched for the third priority words, and the lowest search weight is matched for the fourth priority words.

[0105] The retrieval execution module 26 is used to detect whether there are words with active or passive tags in the retrieval requests after setting part-of-speech priority. If there are words with active tags, the module retrieves retrieval statements containing active tags and increases the retrieval weight of retrieval statements containing active tags. If there are words with passive tags, the module retrieves retrieval statements containing passive tags and increases the retrieval weight of retrieval statements containing passive tags. Active tags are configured for passively received words, and passive tags are configured for actively identified words.

[0106] In this embodiment, active tags are configured for passively received words, and passive tags are configured for actively identified words; the active tags are manually defined attributes that operators extract and summarize based on the characteristics of the search results. The passive tags are the inherent attributes of the search results themselves.

[0107] In one embodiment, specifically, when the search request after setting part-of-speech priority contains words with active tags, the search execution module 26 increases the search weight of the search statement containing active tags by: weighting the search weight of the matched active tags when the active tags are retrieved; when the search request after setting part-of-speech priority contains words with passive tags, the search execution module 26 increases the search weight of the search request containing passive tags by: weighting the search weight of the matched passive tags when the passive tags are retrieved.

[0108] Specifically, the retrieval execution module 26 weights the retrieval weights of the matched active tags by multiplying the retrieval weights of the matched active tags by the highest retrieval weight of the first priority word.

[0109] The retrieval execution module 26 weights the retrieval weights of the matched passive tags by multiplying the retrieval weights of the matched passive tags by the secondary higher retrieval weights of the second priority words.

[0110] The retrieval execution module 26 is also used to prioritize displaying the corresponding retrieval statements according to the size of the increased retrieval weight, and call the search engine to retrieve the retrieval statements in order to obtain the retrieval intent that matches the retrieval request.

[0111] In this embodiment, the retrieval execution module 26 sorts the items by default according to the retrieval weight from largest to smallest, and the earlier the item is listed, the more relevant it is to the user's retrieval intent.

[0112] It should be noted that the division of the various modules in the above system is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing element calls, entirely in hardware, or partially in software calls via processing element calls, with some modules implemented in hardware. For example, module x can be a separate processing element or integrated into a chip within the system. Additionally, module x can be stored as program code in the system's memory, invoked and executed by a processing element. The implementation of other modules is similar. These modules can be fully or partially integrated together or implemented independently. The processing element mentioned here can be an integrated circuit with signal processing capabilities. During implementation, each step of the above method or each of the above modules can be completed through integrated logic circuits in the processor element or through software instructions. These modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), one or more Field Programmable Gate Arrays (FPGAs), etc. When a module is implemented through processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. These modules can be integrated together to form a System-on-a-Chip (SOC).

[0113] Example 3

[0114] In this embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows. Figure 3As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements a business intent retrieval method, a server-side function or step.

[0115] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0116] Receive a search request; the search request contains terms used to search for business intent;

[0117] If the received search request can be broken down, the search request is broken down into several words, and the part of speech of each word is collected and identified.

[0118] Assign part-of-speech priority to each word and match search weights that match the priority to different part-of-speech priorities;

[0119] The system detects whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, the system searches for search statements containing active tags and increases the search weight of search statements containing active tags. If there are words with passive tags, the system searches for search statements containing passive tags and increases the search weight of search statements containing passive tags. Specifically, active tags are configured for passively received words, and passive tags are configured for actively identified words.

[0120] The search query is displayed first according to the weight of the improved search query, and the search engine is called to search the search query to obtain the search intent that matches the search request. The scope of protection of the business intent retrieval method described in this invention is not limited to the execution order of the steps listed in this embodiment. Any solution implemented by adding, subtracting or replacing steps in the prior art based on the principle of this invention is included in the scope of protection of this invention.

[0121] The present invention also provides a business intent retrieval system, which can implement the business intent retrieval method described in the present invention. However, the implementation device of the business intent retrieval method described in the present invention includes, but is not limited to, the structure of the business intent retrieval system listed in this embodiment. All structural modifications and substitutions of the prior art made in accordance with the principles of the present invention are included within the protection scope of the present invention.

[0122] In summary, the business identification system, retrieval method based on the system, storage medium, and device described in this invention analyze the user's true search intent, providing basic tags and part-of-speech classification for the next step of search engine retrieval, thus helping to improve the accuracy of user searches. This invention can obtain the core keywords of user searches and match them with relevant product tags, displaying related products and information, satisfying effective user searches in various business fields, and significantly improving search usage and the rationality of search results. Therefore, this invention effectively overcomes the various shortcomings of existing technologies and has high industrial application value.

[0123] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for retrieving business intent, characterized in that, include: Receive search requests; The search request includes terms used to search for business intent; If the received search request can be broken down, the search request is broken down into several words, and the part of speech of each word is collected and identified. Assign part-of-speech priority to each word and match search weights that match the priority to different part-of-speech priorities; The system detects whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, the system searches for search statements containing active tags and increases the search weight of search statements containing active tags. If there are words with passive tags, the system searches for search statements containing passive tags and increases the search weight of search statements containing passive tags. Active tags are configured for passively received words, and passive tags are configured for actively identified words. The active tags are manually defined attributes that are captured and summarized by the operators based on the characteristics of the search results, and the passive tags are the inherent attributes of the search results. Search queries are displayed in priority according to their increased search weight, and the search engine is invoked to retrieve the search intent that matches the search request.

2. The method for retrieving business intent according to claim 1, characterized in that, After receiving the search request to be retrieved, the business intent retrieval method further includes: if the received search request is indivisible, collecting the indivisible search requests and classifying and storing the collected terms as indivisible terms related to the target industry field and indivisible terms related to the target industry field.

3. The method for retrieving business intent according to claim 2, characterized in that, The method of setting part-of-speech priority for each word and configuring matching search weights for different part-of-speech priorities also includes setting part-of-speech priorities for both divisible and indivisible words.

4. The method for retrieving business intent according to claim 1, characterized in that, The parts of speech of vocabulary include verbs, nouns, gerunds, numerals, and function words; Assigning part-of-speech priorities to each word and matching search weights that correspond to those priorities with different part-of-speech priorities includes: Nouns and noun verbs are set as the first priority vocabulary, verbs and nouns including personal names, geographical terms, organizational terms and proper nouns are set as the second priority vocabulary, adjectives are set as the third priority vocabulary, and numerals are set as the fourth priority vocabulary. The highest search weight is assigned to the first priority words, the next highest search weight is assigned to the second priority words, the lowest search weight is assigned to the third priority words, and the lowest search weight is assigned to the fourth priority words.

5. The method for retrieving business intent according to claim 4, characterized in that, Before setting part-of-speech priorities for each word and matching retrieval weights that match the priorities for different part-of-speech priorities, the retrieval method for the business intent also includes filtering the split function words.

6. The method for retrieving business intent according to claim 4, characterized in that, The steps of retrieving search statements containing active tags and increasing the search weight of search statements containing active tags include: when the active tags are retrieved, weighting the search weight of the matched active tags; The steps of retrieving search statements containing passive tags and increasing the search weight of search statements containing passive tags include: when retrieving the passive tags, weighting the search weight of the matched passive tags.

7. The method for retrieving business intent according to claim 6, characterized in that, The steps for weighting the search weights of the matched active tags include: multiplying the search weight of the active tags by the highest search weight of the first priority word; The steps for weighting the search weights of matched passive tags include multiplying the search weight of the passive tag by the next higher search weight of the second priority word match.

8. A business intent retrieval system, characterized in that, include: The receiving module is used to receive search requests to be retrieved; The search request includes words used to express the search intent; The language processing module is used to, if the received search request can be split, split the search request into several words, collect them and identify the part of speech of each word; The priority setting module is used to set the part-of-speech priority for each word and match the search weight that matches the priority for different parts of speech. The retrieval execution module is used to detect whether there are words with active or passive tags in the search requests after setting part-of-speech priority. If there are words with active tags, the module retrieves search statements containing active tags and increases their search weight. If there are words with passive tags, the module retrieves search statements containing passive tags and increases their search weight. The module prioritizes and displays the corresponding search statements according to their increased search weights, and calls the search engine to retrieve the search statements to obtain search intents that match the search request. Specifically, active tags are configured for passively received words, and passive tags are configured for actively identified words. The active tags are manually defined attributes that are captured and summarized by operators based on the characteristics of the search results, and the passive tags are the inherent attributes of the search results.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method for retrieving business intent as described in any one of claims 1 to 7.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by a processor, it implements the steps of the method for retrieving business intent as described in any one of claims 1 to 7.

Images