Information search method and device, equipment and storage medium

Abstract

The invention belongs to the field of computers, and discloses an information search method and device, equipment and a storage medium. The method comprises the following steps: in response to a triggered information search instruction, obtaining at least one search result corresponding to a search keyword in the information search instruction; determining a search intention corresponding to the search keyword; extracting associated information corresponding to the search intention from the search result; and generating and displaying an information abstract based on the associated information. The associated information corresponding to the search intention is extracted from the search result according to the search intention corresponding to the search keyword; and generating and displaying an information abstract based on the associated information. Compared with an existing mode that numerous search results related to the search keywords are directly displayed based on the search keywords, the search results expected by the user can be obtained, and the search efficiency of the user is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an information search method, device, equipment and storage medium. Background Art

[0002] When users use traditional search engines, the search engines will give relevant search results based on the user's search keywords. In order to obtain the expected search answer, the user needs to visit the URL of each search result separately and obtain the expected answer information by viewing the content in each search result. The above search requires the user to actively view each search result related to the keyword in order to obtain the expected answer. The search efficiency is low and the user experience is not good.

[0003] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of the present invention is to provide an information search method, device, equipment and storage medium, aiming to solve the technical problem that the existing search engine can only return search results related to the search keywords, requiring users to select the expected answers from the search results, resulting in low search efficiency.

[0005] To achieve the above object, the present invention provides an information search method, which comprises the following steps:

[0006] In response to the triggered information search instruction, obtaining at least one search result corresponding to the search keyword in the information search instruction;

[0007] Determine the search intent corresponding to the search keyword;

[0008] extracting relevant information corresponding to the search intent from the search results;

[0009] An information summary is generated and displayed based on the associated information.

[0010] Optionally, the step of extracting associated information corresponding to the search intent from the search results includes:

[0011] Generate a semantic vector corresponding to the search result, and establish a content vector index according to the semantic vector;

[0012] Based on the search intent, associated information corresponding to the search intent is extracted from the content vector index.

[0013] Optionally, the step of generating a semantic vector corresponding to the search result and establishing a content vector index according to the semantic vector includes:

[0014] Performing text segmentation on the search results according to a preset text segmentation strategy to obtain a target text;

[0015] A target semantic vector corresponding to the target text is generated, and a content vector index is established according to the target semantic vector.

[0016] Optionally, the step of generating a target semantic vector corresponding to the target text and establishing a content vector index according to the target semantic vector includes:

[0017] Generating a target semantic vector corresponding to the target text;

[0018] Constructing a plurality of clustering spaces based on the target semantic vector, and determining a cluster center of each clustering space respectively;

[0019] A content vector index is established according to the cluster center and the target semantic vector.

[0020] Optionally, the step of generating and displaying an information summary based on the associated information includes:

[0021] splicing the associated information based on the search keyword to obtain a spliced ​​text;

[0022] An information summary is generated and displayed according to the search keyword and the concatenated text.

[0023] Optionally, the step of generating and displaying an information summary based on the associated information includes:

[0024] generating an information summary based on the associated information;

[0025] Obtaining a search summary template in the information search instruction;

[0026] The information summary is displayed based on the search summary template.

[0027] Optionally, the step of determining the search intent corresponding to the search keyword includes:

[0028] Extracting specific modifiers from the search keywords;

[0029] Determining a search type corresponding to the search keyword based on the specific modifier;

[0030] The search intent corresponding to the search keyword is determined according to the search type.

[0031] Optionally, the step of extracting associated information corresponding to the search intent from the search results includes:

[0032] Obtaining additional identification information in the search results;

[0033] Filter the search results according to the additional identification information to obtain target search results;

[0034] Extract relevant information corresponding to the search intention from the target search results.

[0035] Optionally, the step of extracting associated information corresponding to the search intent from the search results includes:

[0036] Determine a target data source corresponding to the search intent;

[0037] Based on the target data source and the data source corresponding to the search result, associated information corresponding to the search intent is extracted from the search result.

[0038] Optionally, after the step of obtaining at least one search result corresponding to the search keyword in the information search instruction in response to the triggered information search instruction, the step further includes:

[0039] Determine whether the search result is a text file;

[0040] When the search result is not the text file, text conversion is performed on the search result to obtain a search result in text form.

[0041] In addition, to achieve the above-mentioned purpose, the present invention also provides an information search device, the device comprising:

[0042] A response module, configured to obtain at least one search result corresponding to a search keyword in the information search instruction in response to the triggered information search instruction;

[0043] A determination module, used to determine the search intent corresponding to the search keyword;

[0044] An extraction module, used to extract relevant information corresponding to the search intent from the search results;

[0045] The information summary generation module is used to generate and display the information summary based on the associated information.

[0046] In addition, to achieve the above-mentioned purpose, the present invention also proposes an information search device, which includes: a memory, a processor, and an information search program stored in the memory and executable on the processor, wherein the information search program is configured to implement the steps of the information search method described above.

[0047] In addition, to achieve the above-mentioned purpose, the present invention further proposes a storage medium, on which an information search program is stored, and when the information search program is executed by a processor, the steps of the information search method described above are implemented.

[0048] The present invention responds to a triggered information search instruction, obtains at least one search result corresponding to a search keyword in the information search instruction; determines the search intent corresponding to the search keyword; extracts related information corresponding to the search intent from the search results; and generates and displays an information summary based on the related information. Since the present invention extracts related information corresponding to the search intent from the search results based on the search intent corresponding to the search keyword; and generates and displays an information summary based on the related information. Compared to the existing method of directly displaying a large number of search results related to the search keyword based on the search keyword, the above method of the present invention can extract related information corresponding to the search intent from the search results based on the search intent corresponding to the search keyword, and generate an information summary based on the related information for display, thereby obtaining the search results expected by the user and improving the user's search efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the structure of an information search device for a hardware operating environment involved in an embodiment of the present invention;

[0050] Figure 2 This is a flow chart of a first embodiment of the information search method of the present invention;

[0051] Figure 3 A schematic diagram of a flow chart of a second embodiment of the information search method of the present invention;

[0052] Figure 4 A schematic diagram of a flow chart of a third embodiment of the information search method of the present invention;

[0053] Figure 5 It is a structural block diagram of the first embodiment of the information search device of the present invention.

[0054] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0055] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0056] Reference Figure 1 , Figure 1 The present invention is a schematic diagram of the structure of an information search device for a hardware operating environment according to an embodiment of the present invention.

[0057] like Figure 1As shown, the information search device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0058] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the information search device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0059] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and an information search program.

[0060] exist Figure 1 In the information search device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the information search device of the present invention can be set in the information search device, and the information search device calls the information search program stored in the memory 1005 through the processor 1001, and executes the information search method provided by the embodiment of the present invention.

[0061] Based on the above information search device, an embodiment of the present invention provides an information search method, referring to Figure 2 , Figure 2 Schematic diagram of the flow chart of the first embodiment of the information search method of the present invention.

[0062] In this embodiment, the information search method includes the following steps:

[0063] Step S10: In response to the triggered information search instruction, obtaining at least one search result corresponding to the search keyword in the information search instruction.

[0064] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a mobile phone, a tablet computer, a personal computer, etc., or an electronic device or information search device that can realize the above functions. The following takes the information search device as an example to illustrate this embodiment and the following embodiments.

[0065] It should be noted that the information search instruction may be an information search command including a search keyword. The obtaining of at least one search result corresponding to the search keyword in the information search instruction may be based on a traditional search engine to search for information related to the search keyword to obtain at least one search result.

[0066] Furthermore, the search results may be text or audio or video files. In order to accurately extract answers that meet the user's expectations from the search results, after step S10, the following further includes: determining whether the search results are text files;

[0067] When the search result is not the text file, text conversion is performed on the search result to obtain a search result in text form.

[0068] It should be noted that when the search result is not the text file, the search result is converted into text to obtain a search result in text form. If the search result is an audio or video file, the voice information in the audio or video file is converted into text to obtain a search result in text form.

[0069] Step S20: Determine the search intent corresponding to the search keyword.

[0070] It should be noted that the search intent may be the intention or purpose to be expressed by the search keyword input by the user when searching, that is, the type and characteristics of the information, product or service that the user wants to find. The determination of the search intent corresponding to the search keyword may be to identify the search intent corresponding to the search keyword through a preset language model. Among them, the preset language model may be generated by an artificial neural network model trained on a large amount of text data. During the training process, the preset language model can learn to build a language model based on the training data, so that it can understand human language and automatically generate text. Deep learning technologies, such as neural network optimization algorithms and natural language processing technologies, are also used in the training process to improve the quality of output answers.

[0071] Step S30: extracting associated information corresponding to the search intent from the search results.

[0072] It should be noted that the associated information may be information in the search results that is related to the user's search intent. Extracting the associated information corresponding to the search intent from the search results may be extracting the information corresponding to the type of search intent in the search results according to the type of search intent, and using the extracted information as the associated information. For example, the types of search intent include: Information query search intent: The user wants to obtain certain information or knowledge, such as understanding the definition of a noun, obtaining information about a person, or inquiring about information about a country.

[0073] Product-service search intent: Users want to find a specific product or service, such as booking a hotel, buying a new smartphone, etc.

[0074] Problem-solving search intent: Users want to solve a specific problem or issue, such as finding that their phone is broken or looking for a solution to a specific technical problem.

[0075] Interactive search intent: Users want to interact with other users or organizations, such as finding a person’s contact information, joining a chat room or social media group, etc.

[0076] Research search intent: Users want to learn about a specific area or topic to prepare for future decisions or action plans, such as understanding future career prospects or learning a new technology.

[0077] In a specific implementation, the information search device extracts related information corresponding to the search intent from the search results according to the search intent of the search keyword.

[0078] Furthermore, the search results generally contain a lot of commercial advertisements and other content unrelated to the search keywords. In order to improve the user's search experience, the step S30 may include: obtaining additional identification information in the search results;

[0079] Filter the search results according to the additional identification information to obtain target search results;

[0080] Extract relevant information corresponding to the search intention from the target search results.

[0081] It should be noted that the additional identification information may be identification information such as advertisements, sources, links, etc. corresponding to the search results. For example, a search result may be identified as an advertisement, and each search result may contain source information. The search results may be screened according to the additional identification information to obtain target search results, which may be search results with the additional identification information of advertisements deleted from the search results. After the user sets the expected information source, the search results with the information source in the additional identification information not matching the information source expected by the user may also be deleted.

[0082] Furthermore, in order to make the generated information summary more in line with the user's expectations, the step S30 may include: determining the target data source corresponding to the search intent;

[0083] Based on the target data source and the data source corresponding to the search result, associated information corresponding to the search intent is extracted from the search result.

[0084] It should be noted that the target data source may be a data source corresponding to the user's search intent. The extracting of associated information corresponding to the search intent from the search results based on the target data source and the data source corresponding to the search results may be extracting search results whose data sources are consistent with the target data source corresponding to the user's search intent as the associated information.

[0085] Step S40: Generate and display an information summary based on the associated information.

[0086] It should be noted that the generation and display of the information summary based on the associated information may be to extract the most important sentences or phrases from the associated information to form the information summary of the associated information and display it. It may also be to input the associated information into the preset language model to obtain the information summary generated by the preset language model based on the associated information.

[0087] Further, in order to improve the user's reading experience, the step S40 may include: splicing the associated information based on the search keyword to obtain a spliced ​​text;

[0088] An information summary is generated and displayed according to the search keyword and the concatenated text.

[0089] It should be noted that the splicing of the associated information may be performed according to the relevance of the associated information to the search keyword, so that the associated information is spliced ​​into a spliced ​​text whose relevance to the search keyword gradually decreases. Specifically, different weights may be assigned to the source and importance of the associated information, and the associated information may be weighted averaged according to the weights of the source and importance to obtain a weighted value, and the associated information may be spliced ​​according to the weighted value. Alternatively, the associated information may be spliced ​​by feature level fusion to obtain a spliced ​​text. Feature level fusion merges the features extracted from multiple information sources, performs information fusion at the feature level, and then performs the next step of processing. The generation and display of the information summary based on the search keyword and the spliced ​​text may be performed by selecting the text with the highest similarity to the search keyword from the spliced ​​text as the information summary for display.

[0090] Further, in order to improve the user's reading experience, the step S40 may include: generating an information summary based on the associated information;

[0091] Obtaining a search summary template in the information search instruction;

[0092] The information summary is displayed based on the search summary template.

[0093] It should be noted that the search summary template may be a template or method for displaying information summaries set by the user. The user may customize the method for displaying summaries, and then the information search device displays the information summaries based on the search summary template specified by the user, which conforms to the user's reading habits and improves the user experience.

[0094] In response to a triggered information search instruction, this embodiment obtains at least one search result corresponding to a search keyword in the information search instruction; determines a search intent corresponding to the search keyword; extracts related information corresponding to the search intent from the search results; and generates and displays an information summary based on the related information. Since this embodiment extracts related information corresponding to the search intent from the search results based on the search intent corresponding to the search keyword; and generates and displays an information summary based on the related information. Compared to the existing method of directly displaying a large number of search results related to the search keyword based on the search keyword, the above method of this embodiment can extract related information corresponding to the search intent from the search results based on the search intent corresponding to the search keyword, and generate an information summary based on the related information for display, thereby obtaining the search results expected by the user and improving the user's search efficiency.

[0095] refer to Figure 3 , Figure 3 Schematic diagram of the flow chart of the second embodiment of the information search method of the present invention.

[0096] Based on the above first embodiment, in this embodiment, step S30 includes:

[0097] Step S301: Generate a semantic vector corresponding to the search result, and establish a content vector index according to the semantic vector.

[0098] It should be noted that the semantic vector may be a vector that can express the text semantics corresponding to the search result. The generating of the semantic vector corresponding to the search result may be to generate the semantic vectors corresponding to each search result respectively, or to input each search result into a preset language model respectively to obtain the semantic vectors corresponding to each search result output by the preset language model. The semantic vector corresponding to the search result may also be generated by a C&W model, a bag of words model, or based on a convolutional neural network (CNN) or a recurrent neural network (RNN). The establishing of a content vector index according to the semantic vector may be to establish an inverted index of the semantic vector according to the semantic vector to obtain a content vector index corresponding to the semantic vector.

[0099] Furthermore, some search results may have more text content, and in the same search result, there may be text that is highly relevant to the search keyword, as well as other content that is irrelevant to the search keyword, such as advertisements, links, etc. Therefore, in order to improve information search efficiency, the step S301 may include: dividing the search results into texts according to a preset text division strategy to obtain target texts;

[0100] A target semantic vector corresponding to the target text is generated, and a content vector index is established according to the target semantic vector.

[0101] It should be noted that the preset text division strategy may be a pre-defined strategy for dividing the content in the search results, for example, dividing the text in the search results by paragraphs, or dividing by the number of characters, and dividing the search results into a plurality of texts with a preset number of characters. The target text may be the text after the search results are divided.

[0102] In a specific implementation, when the preset text segmentation strategy is segmentation by paragraph, the information search device takes each paragraph in the search result as a target text. For example, if the search result contains 10 paragraphs, the target texts corresponding to the search result are 10, and each paragraph is a target text. Then, the target text is vectorized to obtain a target semantic vector, and an inverted index is established for the target semantic vector to obtain a content vector index.

[0103] Furthermore, in order to efficiently retrieve the associated information corresponding to the search keyword from the search results, the step of generating the target semantic vector corresponding to the target text and establishing the content vector index according to the target semantic vector includes:

[0104] Generating a target semantic vector corresponding to the target text;

[0105] Constructing a plurality of clustering spaces based on the target semantic vector, and determining a cluster center of each clustering space respectively;

[0106] A content vector index is established according to the cluster center and the target semantic vector.

[0107] It should be noted that the construction of multiple clustering spaces based on the target semantic vector can construct the target semantic vector into multiple clustering spaces through conventional clustering means, and determine the cluster center of each clustering space. Specifically, the K-Means clustering algorithm, hierarchical clustering algorithm, DBSCAN clustering algorithm and GMM clustering algorithm, etc. can be used, and this embodiment is not limited here. The establishment of a content vector index based on the cluster center and the target semantic vector can be: determining the target cluster center corresponding to the target semantic vector, and the target cluster center is the cluster center of the cluster space where the target semantic vector is located. Calculate the vector distance between the target semantic vector and each of the cluster centers respectively. The method for calculating the vector distance can use Manhattan distance, Hamming distance, Euclidean distance, Mahalanobis distance, Chebyshev distance, etc., which is not limited here. The vector distance between the target semantic vector and other cluster centers is recorded as the first vector distance. The vector distance between the target semantic vector and the target cluster center is recorded as the second vector distance. If the difference between the first vector distance and the second vector distance is less than the preset vector distance, then determine the target cluster space corresponding to the first vector distance whose difference with the second vector distance is less than the preset vector distance. Use the target semantic vector as the edge vector of the target cluster space. Then count the target semantic vectors contained in each cluster space, and establish a content vector index based on the target semantic vectors contained in each cluster space. Through the above steps, the same target semantic vector may belong to different cluster spaces, thereby improving the accuracy of vector retrieval, wherein the preset vector distance can be a pre-set vector distance, which can be set according to actual needs.

[0108] Step S302: extracting association information corresponding to the search intent from the content vector index based on the search intent.

[0109] It should be noted that in order to improve the efficiency of extracting the associated information corresponding to the search intent from the content vector index, the search intent may be a semantic vector corresponding to the search keyword. The search keyword may be vectorized to obtain a semantic vector corresponding to the search keyword. The extraction of the associated information corresponding to the search intent in the content vector index based on the search intent may be to retrieve a semantic vector whose similarity with the semantic vector corresponding to the search keyword is greater than a preset threshold in the content vector index based on the semantic vector corresponding to the search keyword, and use the search result corresponding to the extracted semantic vector as the associated information. The preset threshold may be a pre-set value, which may be adaptively adjusted according to the number of retrieved semantic vectors.

[0110] This embodiment generates a semantic vector corresponding to the search result, and establishes a content vector index based on the semantic vector; and extracts the associated information corresponding to the search intent from the content vector index based on the search intent. This embodiment generates a semantic vector corresponding to the search result, and establishes a content vector index based on the semantic vector; and extracts the associated information corresponding to the search intent from the content vector index based on the search intent. Search results that meet the user's search intent can be obtained, avoiding the user from checking the search results of traditional search engines one by one to find the expected answer, and improving the user's search efficiency.

[0111] refer to Figure 4 , Figure 4 Schematic diagram of the flow chart of the third embodiment of the information search method of the present invention.

[0112] Based on the above embodiments, in this embodiment, step S20 includes:

[0113] Step S201: extracting specific modifiers from the search keyword.

[0114] It should be noted that the specific modifiers may be keywords that can reflect the user's search intent. For example, when the search keyword is "seaside pictures", the words "pictures" and "sea" may be specific modifiers, and based on the specific modifiers, it can be determined that the user's search intent is to search for pictures related to the sea. When the search keyword is "a happy little thing 500 words", the words "500 words", "happiness" and "things" are specific modifiers that reflect the user's search intent.

[0115] Step S202: determining the search type corresponding to the search keyword based on the specific modifier.

[0116] It should be noted that the search type may include: text, voice, image, video, location, email, etc. The search type corresponding to the search keyword may be determined based on the specific modifier. For example, when the search keyword is "seaside pictures", the words "pictures" and "sea" are specific modifiers. Based on the specific modifier "pictures", the search type may be determined to be a search for pictures. When the search keyword is "a little happy thing 500 words", the word "500 words" is a specific modifier. Based on the specific modifier "500 words", the search type may be determined to be a search for text.

[0117] Step S203: Determine the search intent corresponding to the search keyword according to the search type.

[0118] In the specific implementation, when the search keyword is "seaside pictures", the words "pictures" and "sea" are specific modifiers. According to the specific modifier "pictures", the search type can be determined to be a search for pictures. According to the specific modifier "seaside", the search intent can be determined to be pictures of the seaside. When the search keyword is "a happy little thing 500 words", the word "500 words" is a specific modifier. According to the specific modifier "500 words", the search type can be determined to be a search for text. According to the specific modifiers "happiness" and "things", the search intent can be determined to be a 500-word text about happiness.

[0119] This embodiment extracts specific modifiers from the search keyword; determines the search type corresponding to the search keyword based on the specific modifiers; and determines the search intent corresponding to the search keyword based on the search type. It can accurately identify the user's search intent, and then more accurately extract search results, thereby improving the user's search efficiency.

[0120] Reference Figure 5 , Figure 5 It is a structural block diagram of the first embodiment of the information search device of the present invention.

[0121] like Figure 5 As shown, the information search device proposed in the embodiment of the present invention includes:

[0122] A response module 10, configured to obtain at least one search result corresponding to a search keyword in the information search instruction in response to the triggered information search instruction;

[0123] A determination module 20, used to determine the search intent corresponding to the search keyword;

[0124] An extraction module 30, configured to extract associated information corresponding to the search intent from the search results;

[0125] The information summary generation module 40 is used to generate and display an information summary based on the associated information.

[0126] In response to a triggered information search instruction, this embodiment obtains at least one search result corresponding to a search keyword in the information search instruction; determines a search intent corresponding to the search keyword; extracts related information corresponding to the search intent from the search results; and generates and displays an information summary based on the related information. Since this embodiment extracts related information corresponding to the search intent from the search results based on the search intent corresponding to the search keyword; and generates and displays an information summary based on the related information. Compared to the existing method of directly displaying a large number of search results related to the search keyword based on the search keyword, the above method of this embodiment can extract related information corresponding to the search intent from the search results based on the search intent corresponding to the search keyword, and generate an information summary based on the related information for display, thereby obtaining the search results expected by the user and improving the user's search efficiency.

[0127] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0128] In addition, for technical details that are not described in detail in this embodiment, reference can be made to the information search method provided in any embodiment of the present invention, and will not be repeated here.

[0129] Based on the above-mentioned first embodiment of the information search device of the present invention, a second embodiment of the information search device of the present invention is proposed.

[0130] In this embodiment, the extraction module 30 is further used to generate a semantic vector corresponding to the search result, and to establish a content vector index according to the semantic vector;

[0131] Based on the search intent, associated information corresponding to the search intent is extracted from the content vector index.

[0132] Furthermore, the extraction module 30 is further used to perform text segmentation on the search results according to a preset text segmentation strategy to obtain a target text;

[0133] A target semantic vector corresponding to the target text is generated, and a content vector index is established according to the target semantic vector.

[0134] Furthermore, the extraction module 30 is also used to generate a target semantic vector corresponding to the target text;

[0135] Constructing a plurality of clustering spaces based on the target semantic vector, and determining a cluster center of each clustering space respectively;

[0136] A content vector index is established according to the cluster center and the target semantic vector.

[0137] Furthermore, the information summary generating module 40 is further used to splice the related information based on the search keyword to obtain a spliced ​​text;

[0138] An information summary is generated and displayed according to the search keyword and the concatenated text.

[0139] Furthermore, the information summary generating module 40 is further used to generate an information summary based on the associated information;

[0140] Obtaining a search summary template in the information search instruction;

[0141] The information summary is displayed based on the search summary template.

[0142] Furthermore, the determination module 20 is also used to extract specific modifiers from the search keyword;

[0143] Determining a search type corresponding to the search keyword based on the specific modifier;

[0144] The search intent corresponding to the search keyword is determined according to the search type.

[0145] Furthermore, the extraction module 30 is also used to obtain additional identification information in the search results;

[0146] Filter the search results according to the additional identification information to obtain target search results;

[0147] Extract relevant information corresponding to the search intention from the target search results.

[0148] Furthermore, the extraction module 30 is also used to determine the target data source corresponding to the search intent;

[0149] Based on the target data source and the data source corresponding to the search result, associated information corresponding to the search intent is extracted from the search result.

[0150] Furthermore, the response module 10 is also used to determine whether the search result is a text file;

[0151] When the search result is not the text file, text conversion is performed on the search result to obtain a search result in text form.

[0152] Other embodiments or specific implementations of the information search device of the present invention may refer to the above-mentioned method embodiments, which will not be described in detail here.

[0153] In addition, an embodiment of the present invention further provides a storage medium, on which an information search program is stored. When the information search program is executed by a processor, the steps of the information search method described above are implemented.

[0154] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0155] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0156] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0157] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

[0158] The present invention discloses A1, an information search method, the information search method comprising the following steps:

[0159] In response to the triggered information search instruction, obtaining at least one search result corresponding to the search keyword in the information search instruction;

[0160] Determine the search intent corresponding to the search keyword;

[0161] extracting relevant information corresponding to the search intent from the search results;

[0162] An information summary is generated and displayed based on the associated information.

[0163] A2. The information search method as described in A1, wherein the step of extracting associated information corresponding to the search intention from the search results comprises:

[0164] Generate a semantic vector corresponding to the search result, and establish a content vector index according to the semantic vector;

[0165] Based on the search intent, associated information corresponding to the search intent is extracted from the content vector index.

[0166] A3. In the information search method described in A2, the step of generating a semantic vector corresponding to the search result and establishing a content vector index according to the semantic vector comprises:

[0167] Performing text segmentation on the search results according to a preset text segmentation strategy to obtain a target text;

[0168] A target semantic vector corresponding to the target text is generated, and a content vector index is established according to the target semantic vector.

[0169] A4. In the information search method described in A3, the step of generating a target semantic vector corresponding to the target text and establishing a content vector index according to the target semantic vector comprises:

[0170] Generating a target semantic vector corresponding to the target text;

[0171] Constructing a plurality of clustering spaces based on the target semantic vector, and determining a cluster center of each clustering space respectively;

[0172] A content vector index is established according to the cluster center and the target semantic vector.

[0173] A5. In the information search method according to any one of A1 to A4, the step of generating and displaying an information summary based on the associated information comprises:

[0174] splicing the associated information based on the search keyword to obtain a spliced ​​text;

[0175] An information summary is generated and displayed according to the search keyword and the concatenated text.

[0176] A6. In the information search method according to any one of A1 to A4, the step of generating and displaying an information summary based on the associated information comprises:

[0177] generating an information summary based on the associated information;

[0178] Obtaining a search summary template in the information search instruction;

[0179] The information summary is displayed based on the search summary template.

[0180] A7. In the information search method according to any one of A1 to A4, the step of determining the search intent corresponding to the search keyword comprises:

[0181] Extracting specific modifiers from the search keywords;

[0182] Determining a search type corresponding to the search keyword based on the specific modifier;

[0183] The search intent corresponding to the search keyword is determined according to the search type.

[0184] A8. The information search method as described in any one of A1 to A4, wherein the step of extracting associated information corresponding to the search intention from the search results comprises:

[0185] Obtaining additional identification information in the search results;

[0186] Filter the search results according to the additional identification information to obtain target search results;

[0187] Extract relevant information corresponding to the search intention from the target search results.

[0188] A9. The information search method as described in any one of A1 to A4, wherein the step of extracting associated information corresponding to the search intention from the search results comprises:

[0189] Determine a target data source corresponding to the search intent;

[0190] Based on the target data source and the data source corresponding to the search result, associated information corresponding to the search intent is extracted from the search result.

[0191] A10. The information search method as described in any one of A1 to A4, wherein after the step of obtaining at least one search result corresponding to the search keyword in the information search instruction in response to the triggered information search instruction, the method further comprises:

[0192] Determine whether the search result is a text file;

[0193] When the search result is not the text file, text conversion is performed on the search result to obtain a search result in text form.

[0194] The present invention discloses B11, an information search device, the information search device comprising:

[0195] A response module, configured to obtain at least one search result corresponding to a search keyword in the information search instruction in response to the triggered information search instruction;

[0196] A determination module, used to determine the search intent corresponding to the search keyword;

[0197] An extraction module, used to extract relevant information corresponding to the search intent from the search results;

[0198] The information summary generation module is used to generate and display the information summary based on the associated information.

[0199] B12. The information search device as described in B11, wherein the extraction module is further used to generate a semantic vector corresponding to the search result, and to establish a content vector index according to the semantic vector;

[0200] Based on the search intent, associated information corresponding to the search intent is extracted from the content vector index.

[0201] B13, the information search device as described in B12, the extraction module is further used to divide the search results into texts according to a preset text division strategy to obtain a target text;

[0202] A target semantic vector corresponding to the target text is generated, and a content vector index is established according to the target semantic vector.

[0203] B14, the information search device as described in B13, the extraction module is further used to generate a target semantic vector corresponding to the target text;

[0204] Constructing a plurality of clustering spaces based on the target semantic vector, and determining a cluster center of each clustering space respectively;

[0205] A content vector index is established according to the cluster center and the target semantic vector.

[0206] B15. The information search device as described in any one of B11 to B14, wherein the information summary generation module is further used to splice the related information based on the search keyword to obtain a spliced ​​text;

[0207] An information summary is generated and displayed according to the search keyword and the concatenated text.

[0208] B16. The information search device as described in any one of B11 to B14, wherein the information summary generation module is further used to generate an information summary based on the associated information;

[0209] Obtaining a search summary template in the information search instruction;

[0210] The information summary is displayed based on the search summary template.

[0211] B17. The information search device as described in any one of B11 to B14, wherein the determination module is further used to extract specific modifiers from the search keyword;

[0212] Determining a search type corresponding to the search keyword based on the specific modifier;

[0213] The search intent corresponding to the search keyword is determined according to the search type.

[0214] B18. The information search device as described in any one of B11 to B14, wherein the extraction module is further used to obtain additional identification information in the search results;

[0215] Filter the search results according to the additional identification information to obtain target search results;

[0216] Extract relevant information corresponding to the search intention from the target search results.

[0217] The present invention discloses C19, an information search device, which includes: a memory, a processor, and an information search program stored in the memory and executable on the processor, wherein the information search program is configured to implement the steps of the information search method described in any one of A1 to A10.

[0218] The present invention discloses D20, a storage medium, on which an information search program is stored, and when the information search program is executed by a processor, the steps of the information search method as described in any one of A1 to A10 are implemented.

Claims

1. An information search method, characterized in that: The information search method comprises the following steps: In response to the triggered information search instruction, obtaining at least one search result corresponding to the search keyword in the information search instruction; Determine the search intent corresponding to the search keyword; extracting relevant information corresponding to the search intent from the search results; An information summary is generated and displayed based on the associated information.

2. The information search method according to claim 1, characterized in that: The step of extracting associated information corresponding to the search intent from the search results includes: Generate a semantic vector corresponding to the search result, and establish a content vector index according to the semantic vector; Based on the search intent, associated information corresponding to the search intent is extracted from the content vector index.

3. The information search method according to claim 2, characterized in that: The step of generating a semantic vector corresponding to the search result and establishing a content vector index according to the semantic vector includes: Performing text segmentation on the search results according to a preset text segmentation strategy to obtain a target text; A target semantic vector corresponding to the target text is generated, and a content vector index is established according to the target semantic vector.

4. The information search method according to claim 3, characterized in that: The step of generating a target semantic vector corresponding to the target text and establishing a content vector index according to the target semantic vector includes: Generating a target semantic vector corresponding to the target text; Constructing a plurality of clustering spaces based on the target semantic vector, and determining a cluster center of each clustering space respectively; A content vector index is established according to the cluster center and the target semantic vector.

5. The information search method according to any one of claims 1 to 4, characterized in that: The step of generating and displaying an information summary based on the associated information includes: splicing the associated information based on the search keyword to obtain a spliced ​​text; An information summary is generated and displayed according to the search keyword and the concatenated text.

6. The information search method according to any one of claims 1 to 4, characterized in that: The step of generating and displaying an information summary based on the associated information includes: generating an information summary based on the associated information; Obtaining a search summary template in the information search instruction; The information summary is displayed based on the search summary template.

7. The information search method according to any one of claims 1 to 4, characterized in that: The step of determining the search intent corresponding to the search keyword includes: Extracting specific modifiers from the search keywords; Determining a search type corresponding to the search keyword based on the specific modifier; The search intent corresponding to the search keyword is determined according to the search type.

8. An information search device, characterized in that: The information search device comprises: A response module, configured to obtain at least one search result corresponding to a search keyword in the information search instruction in response to the triggered information search instruction; A determination module, used to determine the search intent corresponding to the search keyword; An extraction module, used to extract relevant information corresponding to the search intent from the search results; The information summary generation module is used to generate and display the information summary based on the associated information.

9. An information search device, characterized in that: The device comprises: a memory, a processor, and an information search program stored in the memory and executable on the processor, wherein the information search program is configured to implement the steps of the information search method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores an information search program, and when the information search program is executed by the processor, the steps of the information search method according to any one of claims 1 to 7 are implemented.

Images