Information display method and device, equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an information display method and device, equipment and a storage medium. Search information is obtained, and multiple search dimensions corresponding to the search information are determined, so that selection of a subsequent search database is facilitated; the method comprises the following steps: searching information in a database corresponding to a plurality of search dimensions according to the search information to obtain search results of different dimensions, and feeding back the searched search results to a terminal, so that the terminal displays the search results of each search dimension, thereby enhancing the diversity of the search results and improving the search efficiency. The technical problems that in the prior art, the displayed search result is single, and the user experience is poor are solved, and the user experience is improved.

Description

Technical Field

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

[0002] In traditional technology, when searching for information, the search information is generally matched with the data in the database, but databases of different modes cannot communicate with each other, and data of different modes cannot be transferred. Generally, only results of the same category as the search information can be searched, and search results of the same type as the search information are displayed. The results are single, especially for some pictures or melodies that are difficult to effectively describe in words. The results displayed to users are single, resulting in a poor user experience.

[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 display method, device, equipment and storage medium, aiming to solve the technical problem that the search results displayed in the prior art are relatively single and the user experience is poor.

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

[0006] Get search information;

[0007] Determining multiple search dimensions corresponding to the search information;

[0008] Performing information search in the database corresponding to the multiple search dimensions according to the search information to obtain search results corresponding to each search dimension;

[0009] The search results are fed back to the terminal so that the terminal displays the search results corresponding to each search dimension.

[0010] Optionally, determining the multiple search dimensions corresponding to the search information includes:

[0011] determining a category of the search information;

[0012] Extracting key feature information corresponding to the category from the search information;

[0013] Determine the corresponding search intent based on the key feature information;

[0014] The corresponding multiple search dimensions are determined according to the search intent.

[0015] Optionally, determining the corresponding search intent based on the key feature information includes:

[0016] Get historical search records;

[0017] The search intent corresponding to the key feature information is determined based on the historical search records through a trained feature recognition model.

[0018] Optionally, determining the search intent corresponding to the key feature information by using a trained feature recognition model according to the historical search records includes:

[0019] Get multiple target historical search records within a preset time period;

[0020] Extracting historical attribute information corresponding to each target historical search record;

[0021] Weighting the historical attribute information corresponding to each target historical search record according to a preset weighting rule, wherein the preset weighting rule is determined according to the time interval between the historical search record and the current moment, the number of searches, and the time interval between adjacent search records;

[0022] Calculate the total weight of each historical attribute information in the historical search record of the target;

[0023] The corresponding search intent is determined according to the total weight and the key feature information through a trained feature recognition model.

[0024] Optionally, determining the corresponding search intent based on the key feature information further includes:

[0025] Extracting at least one attribute information contained in the key feature information;

[0026] Calculate similarity between the search information and the standard instances corresponding to each attribute information;

[0027] The search intent is determined based on the similarity calculation result and the standard instance.

[0028] Optionally, the calculating similarities between the search information and standard instances corresponding to each attribute information respectively includes:

[0029] Obtain instance attribute information of each standard instance;

[0030] Matching the attribute information of the search information with the instance attribute information;

[0031] Determine the first attribute information that successfully matches and the second attribute information that failed to match;

[0032] A similarity calculation is performed based on the first attribute information and the second attribute information.

[0033] Optionally, the determining the corresponding multiple search dimensions according to the search intent includes:

[0034] Determining a search intent category corresponding to the search intent;

[0035] Expanding the relevance of the search intent according to the rewriting strategy corresponding to the search intent category to obtain multiple relevant search intents;

[0036] The corresponding search dimension is determined according to the relevant search intent.

[0037] Optionally, searching for information in databases corresponding to the multiple search dimensions according to the search information includes:

[0038] When the search information contains character data, extracting search key characters from the character data;

[0039] Information search is performed in databases corresponding to the multiple search dimensions according to the search key characters.

[0040] Optionally, performing information search in databases corresponding to the multiple search dimensions according to the search information further includes:

[0041] When the search information contains audio data, extracting time domain features and frequency domain features of the audio data;

[0042] Information search is performed in the database corresponding to the multiple search dimensions according to the time domain features and the frequency domain features.

[0043] Optionally, performing information search in databases corresponding to the multiple search dimensions according to the search information further includes:

[0044] When image data exists in the search information, grayscale features of the image data are extracted using a trained image feature extraction model;

[0045] Information search is performed in databases corresponding to multiple search dimensions according to the grayscale features.

[0046] Optionally, before feeding back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension, the method further includes:

[0047] Calculate the feature similarity between each search result;

[0048] sorting the search results according to the feature similarity;

[0049] Selecting a target search result according to the ranking results of each search result;

[0050] Accordingly, feeding back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension includes:

[0051] The target search result is fed back to the terminal so that the terminal synchronously displays the target search result.

[0052] Optionally, feeding back the target search result to the terminal so that the terminal synchronously displays the target search result includes:

[0053] Allocate a display area corresponding to each target search result according to the ranking result of each search result;

[0054] The display area and the target search result are fed back to the terminal, so that the terminal synchronously displays the target search result according to the display area.

[0055] In addition, to achieve the above-mentioned purpose, the present invention also proposes an information display device, which includes:

[0056] An acquisition module, used to acquire search information;

[0057] A determination module, used to determine multiple search dimensions corresponding to the search information;

[0058] A search module, configured to search for information in a database corresponding to the multiple search dimensions according to the search information, and obtain search results corresponding to each search dimension;

[0059] The display module is used to feed back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension.

[0060] Optionally, the determination module is further used to determine the category of the search information;

[0061] Extracting key feature information corresponding to the category from the search information;

[0062] Determine the corresponding search intent based on the key feature information;

[0063] The corresponding multiple search dimensions are determined according to the search intent.

[0064] Optionally, the determination module is further used to obtain historical search records;

[0065] The search intent corresponding to the key feature information is determined based on the historical search records through a trained feature recognition model.

[0066] Optionally, the determination module is further used to obtain multiple target historical search records within a preset time period;

[0067] Extracting historical attribute information corresponding to each target historical search record;

[0068] Weighting the historical attribute information corresponding to each target historical search record according to a preset weighting rule, wherein the preset weighting rule is determined according to the time interval between the historical search record and the current moment, the number of searches, and the time interval between adjacent search records;

[0069] Calculate the total weight of each historical attribute information in the historical search record of the target;

[0070] The corresponding search intent is determined according to the total weight and the key feature information through a trained feature recognition model.

[0071] Optionally, the determination module is further used to extract at least one attribute information included in the key feature information;

[0072] Calculate similarity between the search information and the standard instances corresponding to each attribute information;

[0073] The search intent is determined based on the similarity calculation result and the standard instance.

[0074] Optionally, the determination module is further used to obtain instance attribute information of each standard instance;

[0075] Matching the attribute information of the search information with the instance attribute information;

[0076] Determine the first attribute information that successfully matches and the second attribute information that failed to match;

[0077] A similarity calculation is performed based on the first attribute information and the second attribute information.

[0078] In addition, to achieve the above-mentioned purpose, the present invention also proposes an information display device, which includes: a memory, a processor, and an information display program stored in the memory and executable on the processor, wherein the information display program is configured to implement the steps of the information display method described above.

[0079] In addition, to achieve the above-mentioned purpose, the present invention further proposes a storage medium, on which an information display program is stored, and when the information display program is executed by a processor, the steps of the information display method described above are implemented.

[0080] The present invention obtains search information and determines multiple search dimensions corresponding to the search information to facilitate the subsequent selection of a search database, then performs information search in the databases corresponding to the multiple search dimensions based on the search information to obtain search results of different dimensions, and finally sends the search results to a search result feedback value terminal so that the terminal displays the search results of each search dimension, thereby enhancing the diversity of search results, avoiding the technical problems of the prior art that the search results displayed are relatively single and the user experience is poor, and improving the user's information acquisition surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is a structural schematic diagram of an information display device for a hardware operating environment involved in an embodiment of the present invention;

[0082] Figure 2 It is a flowchart of the first embodiment of the information display method of the present invention;

[0083] Figure 3 It is a flowchart of the second embodiment of the information display method of the present invention;

[0084] Figure 4 A schematic diagram of a display panel according to an embodiment of an information display method of the present invention;

[0085] Figure 5 It is a flowchart of a third embodiment of the information display method of the present invention;

[0086] Figure 6 It is a structural block diagram of the first embodiment of the information display device of the present invention.

[0087] 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

[0088] 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.

[0089] Reference Figure 1 , Figure 1 It is a schematic diagram of the structure of the information display device of the hardware operating environment involved in the embodiment of the present invention.

[0090] like Figure 1As shown, the information display 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.

[0091] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the information display device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0092] 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 display program.

[0093] exist Figure 1 In the information display 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 display device of the present invention can be set in the information display device, and the information display device calls the information display program stored in the memory 1005 through the processor 1001, and executes the information display method provided by the embodiment of the present invention.

[0094] The embodiment of the present invention provides an information display method, referring to Figure 2 , Figure 2 The figure is a flow chart of a first embodiment of an information display method of the present invention.

[0095] In this embodiment, the information display method includes the following steps:

[0096] Step S10: Obtain search information.

[0097] It should be noted that the executor of the method of this embodiment can be a device with functions such as data processing, program running and data transmission, such as: a computer or mobile phone equipped with an intelligent assistant, etc., or other electronic devices or terminal devices that can achieve the same or similar functions. This embodiment does not make specific restrictions on this. In this embodiment and the following embodiments, a computer terminal equipped with an intelligent assistant will be used as an example for explanation.

[0098] It is worth noting that in order to improve information processing capabilities, in some embodiments, an intelligent assistant can be installed on the terminal device. The intelligent assistant can be a program built based on the natural language processing technology of the Generative Pre-trained Transformer (GPT) model. The intelligent assistant can receive the user's input text, parse the input text into a computer-understandable data format, and then use the pre-trained neural network model to generate response text.

[0099] The terminal in this embodiment can be installed with an intelligent assistant software program, and can also be equipped with devices with functions such as voice acquisition and image acquisition. The input method of input information includes but is not limited to: voice input, text input and data import, etc. Depending on the input method, the search information can be a combination of one or more of audio information, picture information, text information and data table forms, and this embodiment does not make specific restrictions on this.

[0100] Step S20: Determine multiple search dimensions corresponding to the search information.

[0101] It should be noted that the search dimensions correspond to the categories of search information, and different dimensions have corresponding search databases that can store a large amount of data. The search dimensions include but are not limited to: image dimension, audio dimension, and text dimension, etc.

[0102] In specific implementations, different types of search information can generally only correspond to search dimensions with higher category relevance. When searching data across modalities, the search results are often not accurate enough. For example, when the search information is a segment of audio, the search dimension corresponds to the song dimension. The basic information of the audio can be obtained through query, such as the song name, singer, etc., but the audio's simple score or record information will not be found.

[0103] Step S30: performing information search in databases corresponding to the multiple search dimensions according to the search information, and obtaining search results corresponding to each search dimension.

[0104] It can be understood that the databases corresponding to different dimensions include but are not limited to image retrieval databases, audio retrieval databases and text retrieval databases, among which the image retrieval database includes a content-based image retrieval database (CBIR) and a tag-based image retrieval database (ABIR), such as Caltech 101, ImageNet and CIFAR databases; the audio retrieval database includes a content-based audio retrieval database (CBAR) and a tag-based audio retrieval database (ABAR), such as FMA, GTZAN and MagnaTagATune databases; the text retrieval database includes a relational database, a non-relational database and a graph database, such as MySQL or MongoDB databases, which are not specifically limited in this embodiment.

[0105] In a specific implementation, since feature extraction is performed on different types of search information, the attributes of some search information will be weakened and the accuracy of the search will be reduced. Different strategies can be selected according to the category of the search information to implement information search and improve the accuracy of information search. That is, when character data exists in the search information, the search keyword characters in the character data are extracted; and information search is performed in the database corresponding to the multiple search dimensions based on the search keyword characters.

[0106] The extraction of the search key characters can be determined based on whether the characters are special symbols, the part of speech of the character group, and the position in the character group of the entire sentence.

[0107] When the search information contains audio data, extracting time domain features and frequency domain features of the audio data;

[0108] Information search is performed in the database corresponding to the multiple search dimensions according to the time domain features and the frequency domain features.

[0109] It can be understood that the time domain features refer to the amplitude, frequency and phase of the sound signal, while the frequency domain features are mainly the results of spectrum analysis of the sound signal.

[0110] When image data exists in the search information, grayscale features of the image data are extracted using a trained image feature extraction model;

[0111] Information search is performed in databases corresponding to multiple search dimensions according to the grayscale features.

[0112] The image feature extraction model in this embodiment can be a model built based on a convolutional neural network algorithm, or other models that can achieve the same or similar functions. This embodiment does not impose any specific restrictions on this.

[0113] Among them, grayscale generally refers to the brightness value of each pixel in the image.

[0114] Step S40: Feedback the search results to the terminal, so that the terminal displays the search results corresponding to each search dimension.

[0115] In a specific implementation, the search information "apple" is taken as an example for explanation. For "apple", the corresponding search dimensions can be picture dimensions, text dimensions, audio and video dimensions, and link dimensions, etc. The corresponding display results are apple pictures of Apple mobile phones or fruits, apple-related introductions, and related music of Apple mobile phones, such as: power on and off music, purchase link websites, etc., and then the searched information is synchronously displayed on the display panel of the terminal device. The synchronous display can be displayed together so that the search results are displayed on the same page, or it can be displayed sequentially on the page in a certain order. The user can see all the search results by dragging the page. This embodiment does not impose specific restrictions on this.

[0116] This embodiment obtains search information and determines multiple search dimensions corresponding to the search information to facilitate the subsequent selection of a search database, and then searches for information in databases corresponding to the multiple search dimensions based on the search information to obtain search results of different dimensions. Finally, the search is sent to a search result feedback value terminal so that the terminal displays the search results of each search dimension, thereby enhancing the diversity of search results, avoiding the technical problem that the search results displayed by the prior art are relatively single and the user experience is poor, and improving the user's information acquisition surface.

[0117] refer to Figure 3 , Figure 3 It is a flowchart of a second embodiment of an information display method of the present invention.

[0118] Based on the above first embodiment, in this embodiment, step S20 includes:

[0119] Step S201: Determine the category of the search information.

[0120] It should be noted that the categories of search information include but are not limited to: image category, text category, audio category, video category, data table form and link category, etc. For the sake of convenience of explanation, this embodiment and the following embodiments focus on image category, text category and audio category search information as an example.

[0121] Step S202: extracting key feature information corresponding to the category from the search information.

[0122] It should be understood that when the search information is image-type information, the features of the image data can be extracted through a model built based on a convolutional neural network. The features of the image data may include edges, textures, and colors, etc. This embodiment does not impose any specific restrictions on this.

[0123] When the search information is text information, the character data in the search information can be extracted, and the character data features can be extracted through a bag of words (BoW) model or a term frequency–inverse document frequency (tf-idf) model. The features of the text data include but are not limited to: frequency, length, and vector, etc.; the text data can also be semantically analyzed, wherein the semantic analysis includes: lexical analysis, syntactic analysis, and paragraph analysis, etc., and this embodiment does not impose specific restrictions on this.

[0124] When the search information is audio data, the time domain features and frequency domain features of the audio data are generally calculated, where the time domain features refer to the amplitude, frequency and phase of the sound signal, while the frequency domain features are mainly the results of spectral analysis of the sound signal.

[0125] For example, when the search information is the text information of "yellow fruit", the corresponding key feature information is "yellow" and "fruit". Figure 4 When the picture shown is a TV, the corresponding key feature information is "TV" or "display", where the display includes: a display frame 1, a support 2 and a camera 3; when the search information is a piece of audio, the key feature information is the parameters such as the amplitude, frequency and phase of the audio at each time point.

[0126] In a specific implementation, extracting key feature information corresponding to the category from the search information includes: performing feature extraction on the search information according to a feature extraction model corresponding to the category of the search information to obtain the key feature information of the search information.

[0127] Step S203: Determine the corresponding search intent based on the key feature information.

[0128] It should be understood that search intent refers to the information that users want to obtain or the problems that they want to solve when using smart assistants for retrieval. In different scenarios, the search intent of the same pendant feature information may be different. For example, when the current input information is apple, and the user's mobile phone is damaged, the possibility that the search intent is the mobile phone is greater; when it is the season for fruits to be on the market, the possibility that the search intent is apples among fruits is greater.

[0129] Further, in order to improve the determination of the corresponding search intent based on the key feature information, it includes:

[0130] Get historical search records;

[0131] The search intent corresponding to the key feature information is determined based on the historical search records through a trained feature recognition model.

[0132] It is understandable that the historical search records can be records of search text, images or audio within a period of time in history, or text input by the terminal within a period of time in history, for example: text information input when the terminal communicates with an external device, or text that is historically input and deleted by the terminal, even if the above text does not trigger the production of a slide show and there is no parameter information search.

[0133] In the specific implementation, the search intent corresponding to the same input information may be different under the background of different historical input information. For example, when the current input information is apple, and the historical input information is mainly pears, peaches, etc., then the search intent corresponding to the input information is fruit-related; if the historical input information is Huawei mobile phones, Xiaomi mobile phones, etc., then the search intent corresponding to the input information is related to mobile phones or communication devices.

[0134] Furthermore, the determining the search intent corresponding to the key feature information by using a trained feature recognition model according to the historical search records includes:

[0135] Get multiple target historical search records within a preset time period;

[0136] Extracting historical attribute information corresponding to each target historical search record;

[0137] Weighting the historical attribute information corresponding to each target historical search record according to a preset weighting rule, wherein the preset weighting rule is determined according to the time interval between the historical search record and the current moment, the number of searches, and the time interval between adjacent search records;

[0138] Calculate the total weight of each historical attribute information in the historical search record of the target;

[0139] The corresponding search intent is determined according to the total weight and the key feature information through a trained feature recognition model.

[0140] It is worth noting that the historical attribute information corresponding to the historical search record refers to the identification characteristics of the search record. For example, for the historical search record "apple", its attribute information can be apple among fruits or Apple mobile phone, and its attributes are completely different.

[0141] In a specific implementation, the shorter the time interval between the historical search record and the current moment, the higher the score assigned; the more times the same search record is searched, the higher the score assigned to the search record; the shorter the time interval between adjacent search records, the higher the score assigned to the search record. Among them, the attribute information of a single search record is weighted, and the value can be compared and assigned according to the category of the pre-search record and the data marked in the corresponding database. This embodiment does not impose specific restrictions on this.

[0142] In a specific implementation, for a search record, the greater the weight of a certain attribute information, the closer the corresponding search intent is to the dimension corresponding to the attribute information.

[0143] Among them, the preset duration can be set by the user, and the intelligent assistant can also make adjustments according to the actual situation to improve the accuracy of confirming the search intention.

[0144] Furthermore, this embodiment may also provide another method for determining search intent, wherein determining the corresponding search intent based on the key feature information further includes:

[0145] Extracting at least one attribute information contained in the key feature information;

[0146] Calculate similarity between the search information and the standard instances corresponding to each attribute information;

[0147] The search intent is determined based on the similarity calculation result and the standard instance.

[0148] In the specific implementation, the standard instances corresponding to each attribute include but are not limited to the pictures, introductions and connections marked under the attribute. Taking "Apple mobile phone" as an example, the corresponding standard instances can also be pictures of various models of Apple mobile phones or prices and function introductions of various models of Apple mobile phones. This embodiment does not impose specific restrictions on this.

[0149] It is understandable that any data or information is composed of multiple attributes, but the weight of each attribute is different. By matching the number of attributes of the standard instance with the search information, the similarity between the two can be effectively judged, and similarity calculation can be achieved.

[0150] Step S204: Determine the corresponding multiple search dimensions based on the search intent.

[0151] It should be noted that search intent refers to the information that users want to obtain or the problems that they want to solve when searching with smart assistants. In the context of different historical input information, the search intent corresponding to the same input information may be different. For example, when the current input information is apples, and the historical input information mainly includes information such as pears and peaches, then the search intent corresponding to the input information is related to fruits, and the corresponding search dimension is the dimension related to fruits; if the historical input information is Huawei mobile phones, Xiaomi mobile phones, etc., then the search intent corresponding to the input information is related to mobile phones or communication devices, and the corresponding search dimension is the dimension related to electronic devices.

[0152] Furthermore, in order to enrich the search results, the corresponding multiple search dimensions are determined according to the search intent, including:

[0153] Determining a search intent category corresponding to the search intent;

[0154] Expanding the relevance of the search intent according to the rewriting strategy corresponding to the search intent category to obtain multiple relevant search intents;

[0155] The corresponding search dimension is determined according to the relevant search intent.

[0156] It can be understood that the search intent categories include: navigational search, informational search and transactional search, among which navigational search refers to the user's search in order to enter a specific website; informational search is mainly used to find topics, answers and people's information, and can be divided into direct, indirect, suggestion, positioning and list types; transactional search is mainly used for transactions or to query transaction-related information.

[0157] In a specific implementation, when the search intent is a navigational search, you can preview the target website you want to enter, determine the main business direction of the website, and based on the main business direction, find a network link with a corresponding direction through the yellow pages or other resource libraries.

[0158] When the search intent is informational search, the search information can be rewritten to obtain synonyms of the search information, so that the database during the search can be updated and the search content can be expanded. When the search intent is transactional search, products in the same field can be provided, and basic information of the products in the same field can be obtained, and the relevant search intent can be expanded with the basic information. For example, when a user searches for a mobile phone, mobile phone cases, tempered films and other products in the same field are selected, and relevant search intents are generated based on these mobile phone cases, tempered films and other products, and then the corresponding search dimensions are determined.

[0159] This embodiment extracts key feature information of the search information to facilitate determining the search intent of the search information, and then determines multiple search dimensions, thereby increasing the diversity of search results and providing users with a better user experience.

[0160] refer to Figure 5 , Figure 5 It is a flowchart of a third embodiment of an information display method of the present invention.

[0161] Based on the above second embodiment, in this embodiment, before step S40, the following is further included:

[0162] Step S310: Calculate the feature similarity between the search results.

[0163] It should be noted that since there may be multiple types of search results, such as pictures, text introductions, and songs, etc., in order to facilitate subsequent display and to focus on the needs of users, the feature similarity of the search results can be calculated. In this embodiment, the feature similarity calculation between two search results will use one search result as a reference and the other as a comparison. Due to the attribute differences between different search results, the order of similarity calculation between the two will affect the result of the similarity calculation. For example, the text "display panel" and Figure 4 Between the “display panels” shown, when the text “display panel” points to the picture “display panel”, the similarity is relatively high because the picture “display panel” includes the attribute of the text “display panel”; and when the picture “display panel” points to the text “display panel”, due to the influence of the image recognition accuracy, the picture “display panel” may include attributes such as display frame 1, support 2 and camera 3, and the similarity is relatively low.

[0164] Therefore, this embodiment can achieve the correspondence between the search results and the search information by calculating the similarity between two search results. The higher the similarity, the closer it is to the search information.

[0165] Step S320: sorting the search results according to the feature similarity.

[0166] Step S330: Select a target search result according to the ranking results of each search result.

[0167] It can be understood that after determining that the target search structure needs to be displayed, the target search results can be fed back to the terminal so that the terminal can synchronously display the target search results. In this process, in order to display better search results in a limited display area, this embodiment can divide the display area of ​​each search result according to the feature similarity sorting results of each search result.

[0168] Furthermore, feeding back the target search result to the terminal so that the terminal synchronously displays the target search result includes:

[0169] Allocate a display area corresponding to each target search result according to the ranking result of each search result;

[0170] The display area and the target search result are fed back to the terminal, so that the terminal synchronously displays the target search result according to the display area.

[0171] In a specific implementation, the display area size when the terminal displays pictures, text introductions and related links can be different. In this embodiment, in order to highlight the display effect, the display area corresponding to the text introduction can be set to be larger than the network link, so that users can quickly preview related content, improve their reading speed, and have a better viewing experience.

[0172] This embodiment calculates the feature similarities between different search results and then sorts the search results, thereby adjusting the display effect of each search result according to the sorting result, thereby improving the user experience when viewing the search results.

[0173] In addition, an embodiment of the present invention further provides a storage medium, on which an information display program is stored. When the information display program is executed by a processor, the steps of the information display method described above are implemented.

[0174] Since the storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought by the technical solutions of the above embodiments, which will not be described one by one here.

[0175] Reference Figure 6 , Figure 6 It is a structural block diagram of the first embodiment of the information display device of the present invention.

[0176] like Figure 6 As shown, the information display device proposed in the embodiment of the present invention includes:

[0177] The acquisition module 10 is used to acquire search information.

[0178] The determination module 20 is used to determine multiple search dimensions corresponding to the search information.

[0179] The search module 30 is used to search for information in the database corresponding to the multiple search dimensions according to the search information, and obtain search results corresponding to each search dimension.

[0180] The display module 40 is used to feed back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension.

[0181] In one embodiment, the determination module 20 is also used to determine the category of the search information; extract key feature information corresponding to the category from the search information; determine the corresponding search intent based on the key feature information; and determine the corresponding multiple search dimensions through the search intent.

[0182] In one embodiment, the determination module 20 is further used to obtain historical search records; and determine the search intent corresponding to the key feature information through a trained feature recognition model based on the historical search records.

[0183] In one embodiment, the determination module 20 is also used to obtain multiple target historical search records within a preset time length; extract historical attribute information corresponding to each target historical search record; weight the historical attribute information corresponding to each target historical search record according to a preset weighting rule, and the preset weighting rule is determined based on the time interval between the historical search record and the current moment, the number of searches, and the time interval between adjacent search records; count the total weight of each historical attribute information in the target historical search record; determine the corresponding search intention based on the total weight and the key feature information through a trained feature recognition model.

[0184] In one embodiment, the determination module 20 is further used to extract at least one attribute information contained in the key feature information; perform similarity calculations on the search information and the standard instances corresponding to each attribute information; and determine the search intent based on the similarity calculation results and the standard instances.

[0185] In one embodiment, the determination module 20 is also used to obtain instance attribute information of each standard instance; match the attribute information of the search information with the instance attribute information; determine the first attribute information that successfully matches and the second attribute information that fails to match; and calculate similarity based on the first attribute information and the second attribute information.

[0186] In one embodiment, the determination module 20 is also used to determine the search intent category corresponding to the search intent; perform relevance expansion on the search intent according to the rewriting strategy corresponding to the search intent category to obtain multiple related search intents; and determine the corresponding search dimension according to the related search intent.

[0187] In one embodiment, the search module 30 is further configured to extract search key characters from the character data when the search information contains character data; and perform information search in the database corresponding to the multiple search dimensions according to the search key characters.

[0188] In one embodiment, the search module 30 is further used to extract time domain features and frequency domain features of the audio data when the search information contains audio data; and perform information search in the database corresponding to the multiple search dimensions based on the time domain features and the frequency domain features.

[0189] In one embodiment, the search module 30 is further used to extract grayscale features of the image data through a trained image feature extraction model when there is image data in the search information; and to search for information in databases corresponding to multiple search dimensions based on the grayscale features.

[0190] In one embodiment, the display module 40 is also used to calculate the feature similarity between each search result; sort the search results according to the feature similarity; select the target search result according to the sorting result of each search result; accordingly, the search results are fed back to the terminal so that the terminal displays the search results corresponding to each search dimension, including: feeding back the target search results to the terminal so that the terminal synchronously displays the target search results.

[0191] In one embodiment, the display module 40 is further used to allocate a display area corresponding to each target search result according to the ranking result of each search result; and feed back the display area and the target search result to the terminal so that the terminal synchronously displays the target search result according to the display area.

[0192] This embodiment obtains search information and determines multiple search dimensions corresponding to the search information to facilitate the subsequent selection of a search database, and then searches for information in databases corresponding to the multiple search dimensions based on the search information to obtain search results of different dimensions. Finally, the search is sent to a search result feedback value terminal so that the terminal displays the search results of each search dimension, thereby enhancing the diversity of search results, avoiding the technical problem that the search results displayed by the prior art are relatively single and the user experience is poor, and improving the user's information acquisition surface.

[0193] It should be understood that, although the various steps in the flowchart in the embodiment of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and it can be performed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and their execution order is not necessarily performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0194] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present invention. In specific applications, technicians in this field can make settings as needed, and the present invention does not limit this.

[0195] 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.

[0196] In addition, for technical details that are not described in detail in this embodiment, reference can be made to the information display method provided in any embodiment of the present invention, and will not be repeated here.

[0197] In addition, 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.

[0198] 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.

[0199] 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 (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0200] 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.

[0201] The present invention discloses A1, an information display method, the information display method comprising:

[0202] Get search information;

[0203] Determining multiple search dimensions corresponding to the search information;

[0204] Performing information search in the database corresponding to the multiple search dimensions according to the search information to obtain search results corresponding to each search dimension;

[0205] The search results are fed back to the terminal so that the terminal displays the search results corresponding to each search dimension.

[0206] A2. The information display method as described in A1, wherein determining the multiple search dimensions corresponding to the search information includes:

[0207] determining a category of the search information;

[0208] Extracting key feature information corresponding to the category from the search information;

[0209] Determine the corresponding search intent based on the key feature information;

[0210] The corresponding multiple search dimensions are determined according to the search intent.

[0211] A3. The information display method as described in A2, wherein determining the corresponding search intent based on the key feature information comprises:

[0212] Get historical search records;

[0213] The search intent corresponding to the key feature information is determined based on the historical search records through a trained feature recognition model.

[0214] A4. The information display method as described in A3, wherein the step of determining the search intent corresponding to the key feature information by using a trained feature recognition model according to the historical search records comprises:

[0215] Get multiple target historical search records within a preset time period;

[0216] Extracting historical attribute information corresponding to each target historical search record;

[0217] Weighting the historical attribute information corresponding to each target historical search record according to a preset weighting rule, wherein the preset weighting rule is determined according to the time interval between the historical search record and the current moment, the number of searches, and the time interval between adjacent search records;

[0218] Calculate the total weight of each historical attribute information in the historical search record of the target;

[0219] The corresponding search intent is determined according to the total weight and the key feature information through a trained feature recognition model.

[0220] A5. The information display method as described in A2, wherein the determining the corresponding search intent based on the key feature information further includes:

[0221] Extracting at least one attribute information contained in the key feature information;

[0222] Calculate similarity between the search information and the standard instances corresponding to each attribute information;

[0223] The search intent is determined based on the similarity calculation result and the standard instance.

[0224] A6. The information display method as described in A5, wherein the similarity calculation between the search information and the standard instances corresponding to each attribute information comprises:

[0225] Obtain instance attribute information of each standard instance;

[0226] Matching the attribute information of the search information with the instance attribute information;

[0227] Determine the first attribute information that successfully matches and the second attribute information that failed to match;

[0228] A similarity calculation is performed based on the first attribute information and the second attribute information.

[0229] A7. The information display method as described in A2, wherein the determining the corresponding multiple search dimensions according to the search intent includes:

[0230] Determining a search intent category corresponding to the search intent;

[0231] Expanding the relevance of the search intent according to the rewriting strategy corresponding to the search intent category to obtain multiple relevant search intents;

[0232] The corresponding search dimension is determined according to the relevant search intent.

[0233] A8. The information display method as described in A1, wherein the information search is performed in the database corresponding to the multiple search dimensions according to the search information, including:

[0234] When the search information contains character data, extracting search key characters from the character data;

[0235] Information search is performed in databases corresponding to the multiple search dimensions according to the search key characters.

[0236] A9. The information display method as described in A8, wherein the information search is performed in the database corresponding to the multiple search dimensions according to the search information, further comprising:

[0237] When the search information contains audio data, extracting time domain features and frequency domain features of the audio data;

[0238] Information search is performed in the database corresponding to the multiple search dimensions according to the time domain features and the frequency domain features.

[0239] A10. The information display method as described in A9, wherein the information search is performed in the database corresponding to the multiple search dimensions according to the search information, further comprising:

[0240] When image data exists in the search information, grayscale features of the image data are extracted using a trained image feature extraction model;

[0241] Information search is performed in databases corresponding to multiple search dimensions according to the grayscale features.

[0242] A11. The information display method as described in any one of A1 to A10, before feeding back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension, further comprising:

[0243] Calculate the feature similarity between each search result;

[0244] sorting the search results according to the feature similarity;

[0245] Selecting a target search result according to the ranking results of each search result;

[0246] Accordingly, feeding back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension includes:

[0247] The target search result is fed back to the terminal so that the terminal synchronously displays the target search result.

[0248] A12. The information display method as described in A11, wherein the feeding back the target search result to the terminal so that the terminal synchronously displays the target search result comprises:

[0249] Allocate a display area corresponding to each target search result according to the ranking result of each search result;

[0250] The display area and the target search result are fed back to the terminal, so that the terminal synchronously displays the target search result according to the display area.

[0251] The present invention also discloses B13, an information display device, the information display device comprising:

[0252] An acquisition module is used to acquire search information;

[0253] A determination module, used to determine multiple search dimensions corresponding to the search information;

[0254] A search module, configured to search for information in a database corresponding to the multiple search dimensions according to the search information, and obtain search results corresponding to each search dimension;

[0255] The display module is used to feed back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension.

[0256] B14. The information display device as described in B13, wherein the determination module is further used to determine the category of the search information;

[0257] Extracting key feature information corresponding to the category from the search information;

[0258] Determine the corresponding search intent based on the key feature information;

[0259] The corresponding multiple search dimensions are determined according to the search intent.

[0260] B15. The information display device as described in B14, wherein the determination module is further used to obtain historical search records;

[0261] The search intent corresponding to the key feature information is determined based on the historical search records through a trained feature recognition model.

[0262] B16. The information display device as described in B15, wherein the determination module is further used to obtain multiple target historical search records within a preset time period;

[0263] Extracting historical attribute information corresponding to each target historical search record;

[0264] Weighting the historical attribute information corresponding to each target historical search record according to a preset weighting rule, wherein the preset weighting rule is determined according to the time interval between the historical search record and the current moment, the number of searches, and the time interval between adjacent search records;

[0265] Calculate the total weight of each historical attribute information in the historical search record of the target;

[0266] The corresponding search intent is determined according to the total weight and the key feature information through a trained feature recognition model.

[0267] B17. The information display device as described in B14, wherein the determination module is further used to extract at least one attribute information included in the key feature information;

[0268] Calculate similarity between the search information and the standard instances corresponding to each attribute information;

[0269] The search intent is determined based on the similarity calculation result and the standard instance.

[0270] B18. The information display device as described in B17, wherein the determination module is further used to obtain instance attribute information of each standard instance;

[0271] Matching the attribute information of the search information with the instance attribute information;

[0272] Determine the first attribute information that successfully matches and the second attribute information that failed to match;

[0273] A similarity calculation is performed based on the first attribute information and the second attribute information.

[0274] The present invention also discloses C19, an information display device, which includes: a memory, a processor, and an information display program stored in the memory and executable on the processor, wherein the information display program is configured to implement the information display method described above.

[0275] The present invention also discloses D20, a storage medium, on which an information display program is stored, and when the information display program is executed by a processor, the information display method as described above is implemented.

Claims

1. An information display method, characterized in that: The information display method comprises: Get search information; Determining multiple search dimensions corresponding to the search information; Performing information search in the database corresponding to the multiple search dimensions according to the search information to obtain search results corresponding to each search dimension; The search results are fed back to the terminal so that the terminal displays the search results corresponding to each search dimension.

2. The information display method according to claim 1, characterized in that: The determining a plurality of search dimensions corresponding to the search information includes: determining a category of the search information; Extracting key feature information corresponding to the category from the search information; Determine the corresponding search intent based on the key feature information; The corresponding multiple search dimensions are determined according to the search intent.

3. The information display method according to claim 2, characterized in that: The determining the corresponding search intent based on the key feature information includes: Get historical search records; The search intent corresponding to the key feature information is determined based on the historical search records through a trained feature recognition model.

4. The information display method according to claim 3, characterized in that: The determining the search intent corresponding to the key feature information by using a trained feature recognition model according to the historical search records includes: Get multiple target historical search records within a preset time period; Extracting historical attribute information corresponding to each target historical search record; Weighting the historical attribute information corresponding to each target historical search record according to a preset weighting rule, wherein the preset weighting rule is determined according to the time interval between the historical search record and the current moment, the number of searches, and the time interval between adjacent search records; Calculate the total weight of each historical attribute information in the historical search record of the target; The corresponding search intent is determined according to the total weight and the key feature information through a trained feature recognition model.

5. The information display method according to claim 2, characterized in that: The determining the corresponding search intent based on the key feature information further includes: Extracting at least one attribute information contained in the key feature information; Calculate similarity between the search information and the standard instances corresponding to each attribute information; The search intent is determined based on the similarity calculation result and the standard instance.

6. The information display method according to claim 5, characterized in that: The calculating similarity between the search information and the standard instances corresponding to the attribute information respectively includes: Obtain instance attribute information of each standard instance; Matching the attribute information of the search information with the instance attribute information; Determine the first attribute information that successfully matches and the second attribute information that failed to match; A similarity calculation is performed based on the first attribute information and the second attribute information.

7. The information display method according to claim 2, characterized in that: The determining of the corresponding multiple search dimensions according to the search intent includes: Determining a search intent category corresponding to the search intent; Expanding the relevance of the search intent according to the rewriting strategy corresponding to the search intent category to obtain multiple relevant search intents; The corresponding search dimension is determined according to the relevant search intent.

8. An information display device, characterized in that: The information display device comprises: An acquisition module is used to acquire search information; A determination module, used to determine multiple search dimensions corresponding to the search information; A search module, configured to search for information in a database corresponding to the multiple search dimensions according to the search information, and obtain search results corresponding to each search dimension respectively; The display module is used to feed back the search results to the terminal so that the terminal displays the search results corresponding to each search dimension.

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

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

Images