Data processing method and device, electronic equipment and computer storage medium

Abstract

Embodiments of the present application provide a data processing method and device, electronic equipment and computer storage medium. The data processing method is used to generate a description information file related to a target object, comprising: obtaining a description framework and a plurality of materials of a plurality of types related to the target object, the description framework comprising attribute selection information corresponding to the target object; performing at least one processing on each material to obtain attribute information of the material, the attribute information comprising an attribute level and attribute content; selecting a target material from the material, the attribute content and the attribute level of which match the attribute selection information of the description framework; and generating the description information file according to the description framework and the target material. The data processing method can automatically generate the description information file.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a data processing method, apparatus, electronic device, and computer storage medium. Background Technology

[0002] With the development of artificial intelligence (AI) technology, its applications are becoming increasingly widespread. For example, AI is used to generate descriptive profiles for target objects, describing their attributes and characteristics. However, due to the large variety and number of target objects, each with its own unique characteristics, the generation of these descriptive profiles still largely relies on manual labor, with AI providing assistance. This makes it difficult to generate corresponding descriptive profiles for a large number of target objects, and results in high labor intensity and costs. Summary of the Invention

[0003] In view of this, embodiments of this application provide a data processing scheme to at least partially solve the above-mentioned problems.

[0004] According to a first aspect of the embodiments of this application, a data processing method is provided for generating a descriptive information file related to a target object, comprising: acquiring a descriptive framework related to the target object and multiple materials of various types, the descriptive framework including attribute selection information corresponding to the target object; performing at least one processing on each material to obtain attribute information of the material, the attribute information including attribute level and attribute content; selecting target materials from the materials whose attribute content and attribute level match the attribute selection information of the descriptive framework; and generating a descriptive information file based on the descriptive framework and the target materials.

[0005] According to a second aspect of the embodiments of this application, a data processing method is provided for generating a description information file related to a target object, comprising: displaying multiple types of materials corresponding to the target object, processing options corresponding to the materials, and description frame editing options on a display interface; receiving a trigger operation on a processing option corresponding to at least one material in the display interface, processing the material to obtain attribute information of the material, and displaying the attribute information, the attribute information including attribute level and attribute content; receiving an editing operation on the description frame editing option, and obtaining the description frame according to the editing result, the description frame including attribute selection information corresponding to the target object; selecting target materials from the materials whose attribute content and attribute level match the attribute selection information of the description frame; and generating a description information file based on the description frame and the target materials.

[0006] According to a third aspect of the embodiments of this application, a data processing apparatus is provided for generating a description information file related to a target object, comprising: a first acquisition module for acquiring a description framework related to the target object and multiple materials of various types, the description framework including attribute selection information corresponding to the target object; a first processing module for performing at least one processing on each material to obtain attribute information of the material, the attribute information including attribute level and attribute content; a first matching module for selecting target materials from the materials whose attribute content and attribute level match the attribute selection information of the description framework; and a first generation module for generating a description information file based on the description framework and the target materials.

[0007] According to a fourth aspect of the embodiments of this application, a data processing apparatus is provided for generating a description information file related to a target object, comprising: a display module for displaying on a display interface multiple types of materials corresponding to the target object, processing options corresponding to the materials, and description frame editing options; a second processing module for receiving a trigger operation on a processing option corresponding to at least one material in the display interface, processing the material to obtain attribute information of the material, and displaying the attribute information, the attribute information including attribute level and attribute content; a receiving module for receiving an editing operation on the description frame editing option, and obtaining the description frame according to the editing result, the description frame including attribute selection information corresponding to the target object; a second matching module for selecting target materials from the materials whose attribute content and attribute level match the attribute selection information of the description frame; and a second generation module for generating a description information file based on the description frame and the target material.

[0008] According to a fifth aspect of the present application, an electronic device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; the memory is used to store at least one executable instruction, wherein the executable instruction causes the processor to perform an operation corresponding to the data processing method described in the first or second aspect.

[0009] According to a sixth aspect of the embodiments of this application, a computer storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the data processing method as described in the first or second aspect.

[0010] According to the data processing scheme provided in the embodiments of this application, a description frame is used to define the framework of the description information file. This decouples the description frame from the source material, allowing the description information file of the target object to be unrestricted by different source materials. By processing the source material to determine attribute information, the target source material is matched according to the description frame, and then the target source material is combined according to the description frame to form the description information file. Because the description frame layers the script according to attributes, it decouples the description frame from the source material, thereby improving adaptability. It can generate corresponding description information files for different target objects, and the description information files can contain the characteristics of different target objects, ensuring the quality of the description information files. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0012] Figure 1A This is a flowchart of the steps of a data processing method according to Embodiment 1 of this application;

[0013] Figure 1B for Figure 1A A schematic diagram of a scenario example in the illustrated embodiment;

[0014] Figure 2A This is a flowchart of the steps of a data processing method according to Embodiment 2 of this application;

[0015] Figure 2B for Figure 2A A schematic diagram of a scenario example in the illustrated embodiment;

[0016] Figure 3 This is a flowchart of the steps of a data processing method according to Embodiment 3 of this application;

[0017] Figure 4 This is a structural block diagram of a data processing apparatus according to Embodiment 4 of this application;

[0018] Figure 5 This is a structural block diagram of a data processing apparatus according to Embodiment 5 of this application;

[0019] Figure 6 This is a schematic diagram of the structure of an electronic device according to Embodiment Six of this application. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art should fall within the protection scope of the embodiments of this application.

[0021] The specific implementation of the embodiments of this application will be further described below with reference to the accompanying drawings.

[0022] Example 1

[0023] Reference Figure 1A The diagram shows a flowchart of the data processing method according to Embodiment 1 of this application.

[0024] In this embodiment, the method includes the following steps:

[0025] Step S102: Obtain a description framework related to the target object and multiple materials of various types. The description framework includes attribute selection information corresponding to the target object.

[0026] The target object can be a commodity, a building, etc., and the commodity can be a physical or virtual product. In this embodiment, the method is used to generate a script for a virtual live stream in a scenario where a virtual object acts as the anchor. Of course, in other embodiments, the method can be applied to any other suitable scenario, and there are no limitations on this.

[0027] The description framework is used to hierarchically categorize the description information files corresponding to the target object according to attributes. This allows different content paragraphs in the generated description information files to describe different attributes of the target object. This enables a type of target object to use the same or similar description frameworks, thereby decoupling the description framework from the target object's material, improving adaptability, reducing costs, and enabling automated generation of description information files.

[0028] Different description frames can be configured for different target objects. In this embodiment, the description frame includes attribute selection information corresponding to the target object. The attribute selection information of the description frame is used to indicate the topic order, the target attribute selected for the topic, and the attribute value selected for the target attribute.

[0029] The topic order includes, for example, product name, highlights, and features. Each topic corresponds to at least one alternative attribute, and each alternative attribute has at least one attribute value. Attribute selection information is also used to indicate the target attribute selected from the alternative attributes corresponding to the topic, and the attribute value selected from the target attribute.

[0030] The following is a descriptive framework for an alcoholic beverage product.

[0031]

[0032] The first row of the table describes the order of the topics involved in the framework, such as product name, highlights, features, and net content. Each topic can be broken down into multiple subtopics, and each subtopic can correspond to one or more attributes. For example, the product name can be broken down into two subtopics, corresponding to the category attribute and the specialty product category attribute, respectively. Highlights can be broken down into two subtopics, corresponding to the other element attribute and the aroma attribute, respectively. Features can be broken down into three subtopics, with two subtopics corresponding to one attribute each (packaging method and flavor), and the other subtopic corresponding to two attributes (ingredients and raw material composition). The attributes corresponding to these topics are the alternative attributes.

[0033] The attribute description and attribute selection in the description frame are used to indicate the target attribute to be selected for the corresponding topic. For example, if the value for the attribute description is "yes" and the value for the attribute selection is "any", it means that some of the candidate attributes for the topic are selected as the target attribute.

[0034] The attribute value selection and attribute value list in the description frame are used to indicate the selected attribute values ​​for the corresponding attributes. For example, if the attribute value list value is "Yes" and the attribute value selection value is "all", it means that all attribute values ​​corresponding to the target attribute are selected.

[0035] The description framework allows control over the content of the description information archive, ensuring that the content is sufficient and meets the requirements without being overly redundant.

[0036] The material may be image-type material and / or text-type material obtained through appropriate means, wherein the text-type material includes at least one of the text obtained by speech recognition of audio-type material and the original text.

[0037] For product-related targets, materials can be obtained from product detail pages, display images, cover images, etc., while text-based materials can be obtained from explanatory text.

[0038] Audio materials can be audio extracted from videos or pure audio without images.

[0039] Step S104: Perform at least one processing on each material to obtain the material's attribute information, which includes attribute hierarchy and attribute content.

[0040] In one feasible approach, image-text recognition is performed on the image-type material to obtain the attribute sentences corresponding to the image-type material. The image-text recognition can employ methods such as neural network models with image recognition capabilities or OCR (Optical Character Recognition) to identify items, text, etc., and convert these items and text into corresponding attribute sentences. Subsequently, attribute information can be determined based on the semantics of the attribute sentences.

[0041] In another feasible approach, for text-type materials, attribute mining can be performed on the text-type materials to obtain the attribute sentences corresponding to the text materials. For example, attribute mining can be performed using natural language processing models capable of semantic recognition to obtain the corresponding attribute sentences, so that candidate materials can determine attribute information based on the semantics of the attribute sentences.

[0042] In this embodiment, the attribute hierarchy of the material is used to indicate the theme corresponding to the material, and the attribute content of the material is used to indicate the attributes and attribute values ​​of the target object described by the material. For example, the attribute content could be "flavor - strawberry", etc.

[0043] Step S106: Select target material from the material whose attribute content and attribute level match the attribute selection information of the description frame.

[0044] In one feasible approach, materials matching the attribute level are determined based on the attribute hierarchy and the theme order indicated in the attribute selection information. Then, materials matching the attribute content are selected from the materials matching the attribute hierarchy based on the selected attributes and attribute values ​​indicated in the attribute selection information.

[0045] For example, if the attribute selection information indicates that some attributes under the topic "Product Name" should be selected for description, and the corresponding attributes for the topic are "Category" and "Specialty Product Category", then the attribute "Category" can be selected as the target attribute. All attribute values ​​under the corresponding "Category" attribute can be selected.

[0046] Accordingly, based on the attribute hierarchy of the materials, materials corresponding to the theme are selected, and materials describing the attribute values ​​in the "category" attribute are selected from these materials as target materials.

[0047] Step S108: Generate a description information file based on the description framework and the target material.

[0048] In one feasible approach, target materials can be combined according to the thematic order in the description framework to generate a description information archive.

[0049] This approach addresses potential semantic repetition and redundancy in source materials. Furthermore, it allows for control over sentence order and content within the descriptive information file based on the descriptive framework, resulting in concise, clear, and free-flowing statements. The decoupling of the descriptive framework from the target object's source materials allows it to be applied to one or more target categories. This enables the automatic generation of corresponding descriptive information files for each target object, and allows for rapid updates to the descriptive information file when the characteristics of the target object change, ensuring excellent applicability.

[0050] Reference Figure 1B This diagram illustrates a descriptive framework in one usage scenario of an embodiment of this application. The materials can be obtained through content mining of the target object's comments, details page, and main image; the method of acquisition is not limited in this usage scenario.

[0051] By processing the obtained materials, their attribute information can be acquired. Attribute information includes attribute levels and attribute content. Attribute levels include, for example, "theme" and "characteristics." Attribute content includes, for example, "category - wine" and "flavor - strawberry."

[0052] Based on the attribute selection information corresponding to the description frame, select the corresponding material from the material, and combine the selected material according to the theme order indicated by the description frame to obtain the description information file.

[0053] Below, using the aforementioned descriptive framework, we provide an example of a descriptive information file for a type of alcoholic beverage. Based on the descriptive framework, the selected attribute under the theme "Characteristics" is "Taste," and all attribute values ​​corresponding to this attribute are selected attribute values. Based on this, we select all attribute values ​​corresponding to "Taste" from the materials at the attribute level of "Features." A similar method can be used to match corresponding materials for other subjects.

[0054] Based on the theme order indicated by the attribute selection information, the materials corresponding to each theme are combined to form a descriptive information archive.

[0055] In this embodiment, a description frame is used to define the framework of the description information file. This decouples the description frame from the source material, allowing the description information file of the target object to be independent of different source materials. By processing the source material to determine its attribute information, the target source material is matched according to the description frame, and then combined with the target source material according to the description frame to form the description information file. Because the description frame layers the script according to attributes, it decouples from the source material, thereby improving adaptability. It can generate corresponding description information files for different target objects, and the description information files can contain the characteristics of different target objects, ensuring the quality of the description information files.

[0056] The data processing method of this embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as mobile phones, PADs, etc.) and PCs.

[0057] Example 2

[0058] Reference Figure 2A The diagram shows a flowchart of the data processing method according to Embodiment 2 of this application.

[0059] In this embodiment, the data processing method includes:

[0060] Step S202: Obtain a descriptive framework and multiple materials of various types related to the target object.

[0061] The descriptive framework includes attribute selection information corresponding to the target object. The descriptive framework can be automatically generated based on the target object, or it can be manually adjusted as needed. By adaptively adjusting the values ​​of attributes or attribute selections for different target objects, a descriptive framework with better matching to the target object can be obtained.

[0062] The attribute selection information of the description framework is used to indicate the topic order, the target attribute selected for the topic, and the attribute value selected for the target attribute.

[0063] The materials for the target object can be obtained in any appropriate way, which will not be elaborated here.

[0064] Step S204: Perform at least one processing on each material to obtain the material's attribute information.

[0065] The types of materials may differ. Appropriate methods can be used to process different types of materials and obtain corresponding attribute sentences in order to obtain attribute information.

[0066] In one feasible approach, step S204 includes the following sub-steps:

[0067] Sub-step S2041: Perform image-text recognition on the image type material to obtain the attribute sentences corresponding to the image type material; and / or, perform attribute mining on the text type material to obtain the attribute sentences corresponding to the text material.

[0068] Image recognition, for example, involves performing object detection and optical character recognition on image-type materials to obtain the text and objects in the image-type materials, and then determining the corresponding attribute sentences based on the text and objects.

[0069] Attribute mining, for example, involves performing natural language processing on text materials to obtain attribute sentences that describe the attributes of a target object.

[0070] Sub-step S2042: Clean the attribute sentences according to their semantic information.

[0071] To reduce the time and computational costs of subsequent matching, improve matching efficiency, and ensure that the generated description information files are semantically clear, concise, and clear, the attribute sentences are cleaned to remove sentences with repetitive meanings, sentences with unclear meanings, and sentences that do not meet the requirements.

[0072] In addition, it can extract summaries from attribute sentences to obtain expert summaries, mine images and text to obtain image and text content, extract information to obtain activity discounts, rewrite detail pages to obtain detail page introductions, and extract comments to obtain short reviews, etc.

[0073] Sub-step S2043: Standardize the attribute content corresponding to the remaining attribute sentences after cleaning, so as to determine the attribute content of the attribute sentences and the corresponding attribute level as the attribute information of the corresponding material based on the standardization result.

[0074] Since the same attribute and attribute value can be described in many different ways, in order to ensure the accuracy of subsequent matching, the attribute content can be standardized to unify the different forms of synonymous expressions into a standard expression.

[0075] Terms like "classification" and "category" can be standardized to "classification". Terms like "taste" and "flavor" can be standardized to "flavor". Attribute values ​​and attribute levels can also be standardized in a similar way.

[0076] The attribute level is determined based on the standardized content corresponding to the "theme," and the attribute content is determined based on the standardized content corresponding to the "attribute" and "attribute value." In this embodiment, the attribute level of the material is used to indicate the theme corresponding to the material, and the attribute content of the material is used to indicate the attributes and attribute values ​​of the target object described by the material.

[0077] Alternatively, the following processing can be performed on the materials corresponding to the attribute level:

[0078] Sub-step S2044: Sort materials with the same attribute value according to the first popularity information, and remove duplicate materials whose semantic similarity meets the set value by using the longest common subsequence algorithm.

[0079] Since the same attribute level can include multiple materials, and the attribute value for the same attribute can have one or more forms of expression, materials with the same attribute value in the same attribute level can be sorted according to the first popularity information.

[0080] The primary trending information can correspond to different metrics depending on the usage scenario. For example, if the object is a product, the primary trending information could be sales popularity. Or, if the object is course content, the primary trending information could be the importance of the knowledge points or the popularity of the knowledge points.

[0081] In addition, the longest common subsequence algorithm can be used to deduplicate the materials, and materials with high semantic similarity can be deduplicated to avoid excessive repetition of semantics and verbose expression.

[0082] Materials with the same attribute at the same attribute level can be sorted according to the second popularity information, which can be the popularity of the selected material, etc.

[0083] Sub-step S2045: Based on the attribute content, form a material group from the remaining materials after removing duplicates of materials with the same attribute content.

[0084] For example, materials with the same attribute can be grouped together. This material group can include different attribute values ​​of the same attribute.

[0085] Sub-step S2046: Combine materials whose attribute content belongs to the same theme into material paragraphs.

[0086] Since the themes correspond to the attribute levels, in order to make the description information archives more logical and the hierarchy clearer, material groups belonging to the same attribute level (i.e., belonging to the same theme) can be grouped into a material paragraph to facilitate the subsequent combination of material paragraphs.

[0087] Step S206: Select target material from the material whose attribute content and attribute level match the attribute selection information of the description frame.

[0088] In one feasible approach, step S206 includes the following sub-steps:

[0089] Sub-step S2061: Based on the theme order indicated by the attribute selection information in the description framework, determine the candidate materials corresponding to the attribute level of each theme.

[0090] Since the attribute hierarchy of the materials corresponds to the theme indicated in the description frame, the attribute hierarchy of the materials can be matched with the theme in the attribute selection information. For example, for the first theme "Product Name" in the description frame, materials with the attribute hierarchy of "Product Name" can be selected as candidate materials.

[0091] Sub-step S2062: Based on the target attributes corresponding to each theme and the attribute values ​​described in the target attribute description, determine the target materials whose attribute content matches the target attributes from the candidate materials.

[0092] Based on the attribute description and attribute selection in the description framework, the target attribute corresponding to the topic can be determined. Based on the attribute value selection and attribute value listing in the description framework, the attribute value corresponding to the target attribute can be determined.

[0093] If the attribute description for the topic "Features" is "Yes" and the attribute selection is "any", then its target attribute can be a subset selected from "Packaging Method", "Ingredients", "Raw Material Ingredients", and "Flavor". For example, one can be selected from each corresponding subtopic, so the target attribute could be "Packaging Method", "Ingredients", and "Flavor".

[0094] Since the attribute value list is set to "Yes" and the attribute value selection is "all", all corresponding attribute values ​​for the target attribute can be listed. Based on this sub-step S2062, it is achieved through the following process:

[0095] Process A1: Based on the target attributes corresponding to each of the topics, select candidate materials whose attribute content matches the target attributes.

[0096] Taking the target attributes as "packaging method", "ingredients", and "flavor" as an example, based on the attribute content of the candidate materials, select all candidate materials whose attribute content contains "packaging method", "ingredients", and "flavor" as matching candidate materials.

[0097] Process A2: Based on the attribute value selected by the target attribute, select candidate materials containing the attribute value selected by the target attribute from the matched candidate materials as the target material.

[0098] For example, if we list all the attribute values ​​corresponding to the target attribute, we can use all matching candidate materials as the target material. Alternatively, if we only list a portion of the target attribute, we can select candidate materials whose attribute content contains the selected attribute value from the matching candidate materials as the target material.

[0099] Step S208: Generate a description information file based on the description framework and the target material.

[0100] In one feasible approach, step S208 can be implemented as follows: sorting the material paragraphs according to the topic order indicated by the description framework and the attribute level corresponding to the material paragraphs formed by the target material, and generating the description information file according to the sorting result.

[0101] Since each material paragraph has a corresponding attribute level, the order of the material paragraphs is determined according to the correspondence between the attribute level and the topics in the description framework and the order of the topics. The material paragraphs are then merged in order to form a description information archive.

[0102] Optionally, to make the information description in the description file smoother and more coherent, the description frame's attribute options also include transition sentence information. This transition sentence information indicates whether to add a transition sentence to or within a source paragraph. If the transition sentence information value is "yes," a transition sentence is added; otherwise, if it is "no," a transition sentence is not added.

[0103] The following is combined Figure 2B Here is an example of a process for generating information archives.

[0104] like Figure 2B As shown, if the target object is a product, then image-type materials are obtained from the main image or details image of the target object, and text-type materials are obtained from the attribute description, selling point description, and explanatory text of the target object.

[0105] For image-based materials, images and OCR text are obtained through OCR recognition, object detection, and content mining, and then corresponding attribute sentences are output based on the OCR text.

[0106] For text-based materials, attribute sentences are obtained through attribute mining (such as XG generation) and combined with existing attribute knowledge.

[0107] For these attribute sentences, sentences highlighting key attributes can be merged and semantically repetitive sentences removed to create content highlighting key attributes. Sentences describing products can be cleaned and processed to create content describing products at that level. Sentences summarizing influencers can be extracted to obtain content summarizing influencers at that level. Similarly, attribute sentences can be processed through image and text mining, information extraction, detail page rewriting, and comment extraction to obtain corresponding content. During the synthesis process, connecting sentences can be added between adjacent paragraphs as needed to make the descriptive information flow more smoothly, avoiding awkwardness and conforming to expression habits for greater coherence and fluency.

[0108] By employing the above methods, using an attribute-based hierarchical description framework and connecting phrases to splice and integrate the constructed materials, semantic integration of multi-source materials can be supported, avoiding issues such as irrelevance, redundancy, and incoherence between content, while maintaining chapter-level richness in content. Multi-modal presentation can effectively enrich the content presentation formats required for virtual live streaming. The description framework is reasonably constructed and has good scalability, allowing for rapid reuse in new categories or industries.

[0109] This method can be applied to the field of virtual live streaming to quickly generate corresponding live streaming scripts. Of course, it can also be applied to other fields to generate the required descriptive information files.

[0110] In this embodiment, a description framework is used to obtain materials corresponding to different attributes from existing materials. The materials are then processed based on attribute selection information within the description framework to obtain a description information file. Because the description framework layers the description information file according to attributes, it decouples the description framework from the existing materials, thereby improving adaptability. It can generate corresponding description information files for different target objects while preserving the characteristics of different target objects, ensuring effective use.

[0111] The data processing method of this embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as mobile phones, PADs, etc.) and PCs.

[0112] Example 3

[0113] Reference Figure 3 The diagram illustrates the steps of the data processing method in this embodiment.

[0114] In this embodiment, the method is used to generate a descriptive information file related to a target object, which includes:

[0115] Step S302: Display various types of materials corresponding to the target object, processing options for the materials, and description frame editing options on the display interface.

[0116] To facilitate obtaining more suitable description frames for different target objects and to classify materials more accurately, the processing options and description frame editing options corresponding to the materials can be displayed on the interface.

[0117] The processing options are used to select the processing methods for the materials. The attribute information obtained after processing can also be displayed on the display interface and viewed and / or edited manually.

[0118] The description frame editing options are used to adjust the description frame, such as modifying the content within it.

[0119] In this embodiment, the description frame editing options include a topic adjustment option for adjusting the order of topics in the description frame, an attribute adjustment option for adjusting the target attribute corresponding to the topic, and an attribute value adjustment option for adjusting the attribute value of the target attribute description.

[0120] The topic adjustment options allow you to add or remove topics as needed, or adjust the order of topics.

[0121] The attribute adjustment options are used to adjust the selected attribute as needed. Similarly, the attribute value adjustment options are used to adjust the selected attribute value as needed.

[0122] Optionally, the description frame editing options may also include a transition sentence adjustment option, which allows you to adjust whether to add transition sentences as needed.

[0123] Step S304: Receive a trigger operation for processing options corresponding to at least one material in the display interface, process the material to obtain the attribute information of the material, and display the attribute information, which includes attribute level and attribute content.

[0124] For example, by triggering processing options, different types of materials can be processed accordingly. For instance, object detection can be performed on image materials to obtain the corresponding attribute sentences and the attribute information corresponding to those attribute sentences, such as attribute level, attribute, and attribute value.

[0125] Attribute mining is performed on text-based materials to obtain corresponding attribute sentences and attribute information such as attribute level, attribute, and attribute value.

[0126] The acquired attribute information can be displayed in the interface for viewing and editing.

[0127] Step S306: Receive the editing operation of the description frame editing options, and obtain the description frame according to the editing result. The description frame includes attribute selection information corresponding to the target object.

[0128] The display interface provides corresponding description framework editing options for each theme, attribute selection, etc. By editing these options, the corresponding values ​​can be changed, thereby obtaining the description framework.

[0129] In one feasible approach, step S306 includes the following sub-steps:

[0130] Sub-step S3061: Receive an adjustment operation on at least one of the theme adjustment option, the attribute adjustment option, and the attribute value adjustment option.

[0131] For example, if you need to add a new theme, you can add it in the theme row through the theme adjustment options.

[0132] And / or, when it is necessary to select all attribute values ​​for a certain attribute, adjust the corresponding attribute value adjustment option to adjust the selected attribute value to "ALL".

[0133] Sub-step S3062: Based on the adjustment result, determine the topic order of the broadcast instruction, the target attribute selected for the topic, and the attribute value of the target attribute to form the attribute selection information of the description framework.

[0134] The order of the topics can be determined by the order from left to right of the topics displayed on the interface.

[0135] The target attribute can be determined based on the values ​​in the attribute selection and attribute description rows in the display interface.

[0136] The attribute value corresponding to the target attribute can be determined based on the values ​​in the two rows of attribute value selection and attribute value list in the display interface.

[0137] Step S308: Select target material from the material whose attribute content and attribute level match the attribute selection information of the description frame.

[0138] The process of matching based on the description framework and materials is similar to that in the aforementioned Embodiment 1 or 2, and therefore will not be described again.

[0139] Step S310: Generate a description information file based on the description framework and the target material.

[0140] The process of generating the description information file is similar to that of the aforementioned Embodiment 1 or 2, and therefore will not be described again.

[0141] Through this embodiment, the description framework and materials can be processed as needed to obtain a more appropriate description framework, thereby ensuring that the generated description information file has a better effect and can meet the requirements.

[0142] The data processing method of this embodiment can be executed by any suitable electronic device with data processing capabilities, including but not limited to: servers, mobile terminals (such as mobile phones, PADs, etc.) and PCs.

[0143] Example 4

[0144] Reference Figure 4 The diagram shows a structural block diagram of the data processing apparatus according to Embodiment 4 of this application.

[0145] In this embodiment, the data processing device is used to generate a descriptive information file related to a target object, which includes:

[0146] The first acquisition module 402 is used to acquire a description framework related to the target object and multiple materials of various types. The description framework includes attribute selection information corresponding to the target object.

[0147] The first processing module 404 is used to perform at least one processing on each material to obtain the material's attribute information, the attribute information including attribute level and attribute content;

[0148] The first matching module 406 is used to select target materials from the materials whose attribute content and attribute level match the attribute selection information of the description frame;

[0149] The first generation module 408 is used to generate a description information file based on the description framework and the target material.

[0150] Optionally, the multiple types of materials include image-type materials and / or text-type materials, wherein the text-type materials include at least one of the text obtained by speech recognition of audio-type materials and the original text; the first processing module 404 is used to perform image-text recognition on the image-type materials to obtain the attribute sentences corresponding to the image-type materials; and / or to perform attribute mining on the text-type materials to obtain the attribute sentences corresponding to the text materials.

[0151] Optionally, the first processing module 404 is further configured to perform cleaning processing on the attribute sentences according to the semantic information of the attribute sentences; and to perform standardization processing on the attribute content corresponding to the remaining attribute sentences after cleaning, so as to determine the attribute content of the attribute sentences and the corresponding attribute levels as the attribute information of the corresponding materials based on the standardization processing results.

[0152] Optionally, the attribute selection information of the description framework is used to indicate the topic order, the target attribute selected for the topic, and the attribute value selected for the target attribute.

[0153] Optionally, the attribute hierarchy of the material is used to indicate the theme corresponding to the material, and the attribute content of the material is used to indicate the attributes and attribute values ​​of the target object described by the material.

[0154] Optionally, the first matching module 406 is used to determine candidate materials corresponding to the attribute levels of each topic according to the topic order indicated by the attribute selection information in the description framework; and to determine target materials whose attribute content matches the target attribute from the candidate materials according to the target attribute corresponding to each topic and the attribute value selected by the target attribute.

[0155] Optionally, the first matching module 406 is used to, when determining target materials whose attribute content matches the target attribute from the candidate materials based on the target attribute corresponding to each theme and the attribute value selected by the target attribute, select candidate materials whose attribute content matches the target attribute based on the target attribute corresponding to each theme; and select candidate materials containing the attribute value selected by the target attribute from the matched candidate materials as the target materials based on the attribute value selected by the target attribute.

[0156] Optionally, the first processing module 404 is further configured to sort materials with the same attribute value according to the first popularity information, and remove duplicate materials whose semantic similarity meets the set value through the longest common subsequence algorithm; according to the attribute content, form material groups by removing duplicate materials with the same attribute content; and combine materials whose attribute content belongs to the same theme into material paragraphs.

[0157] Optionally, the first generation module 408 is used to sort the material paragraphs according to the theme order indicated by the description framework and the attribute level corresponding to the material paragraphs formed by the target material, and generate the description information file according to the sorting result.

[0158] The data processing apparatus of this embodiment is used to implement the corresponding data processing methods in the foregoing multiple method embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here. Furthermore, the functional implementation of each module in the data processing apparatus of this embodiment can be referred to the description of the corresponding part in the foregoing method embodiments, which will also not be repeated here.

[0159] Example 5

[0160] Reference Figure 5 The diagram shows a structural block diagram of the data processing apparatus according to Embodiment 5 of this application.

[0161] In this embodiment, the data processing device is used to generate a descriptive information file related to a target object, which includes:

[0162] The display module 502 is used to display on the display interface various types of materials corresponding to the target object, processing options for the materials, and description frame editing options;

[0163] The second processing module 504 is used to receive a trigger operation for processing options corresponding to at least one material in the display interface, process the material to obtain the attribute information of the material, and display the attribute information, wherein the attribute information includes attribute level and attribute content.

[0164] The receiving module 506 is used to receive the editing operation of the description frame editing options, and obtain the description frame according to the editing result. The description frame includes attribute selection information corresponding to the target object.

[0165] The second matching module 508 is used to select target materials from the materials whose attribute content and attribute level match the attribute selection information of the description frame;

[0166] The second generation module 510 is used to generate a description information file based on the description framework and the target material.

[0167] Optionally, the description frame editing options include a topic adjustment option for adjusting the order of topics in the description frame, an attribute adjustment option for adjusting the target attribute corresponding to the topic, and an attribute value adjustment option for adjusting the attribute value of the target attribute description.

[0168] Optionally, the receiving module 506 is configured to receive adjustment operations on at least one of the topic adjustment options, the attribute adjustment options, and the attribute value adjustment options; and, based on the adjustment results, determine the attribute selection information that forms the description frame by the topic order indicated by the broadcast, the target attribute selected for the topic, and the attribute value described by the target attribute.

[0169] The data processing apparatus of this embodiment is used to implement the corresponding data processing methods in the foregoing multiple method embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here. Furthermore, the functional implementation of each module in the data processing apparatus of this embodiment can be referred to the description of the corresponding part in the foregoing method embodiments, which will also not be repeated here.

[0170] Example 6

[0171] Reference Figure 6 The diagram shows a structural schematic of an electronic device according to Embodiment Six of this application. The specific embodiments of this application do not limit the specific implementation of the electronic device.

[0172] like Figure 6 As shown, the electronic device may include: a processor 602, a communication interface 604, a memory 606, and a communication bus 608.

[0173] in:

[0174] The processor 602, communication interface 604, and memory 606 communicate with each other via communication bus 608.

[0175] Communication interface 604 is used for communication with other electronic devices or servers.

[0176] The processor 602 is used to execute program 610, specifically to perform the relevant steps in the above data processing method embodiments.

[0177] Specifically, program 610 may include program code that includes computer operation instructions.

[0178] The processor 602 may be a central processing unit (CPU), a specific integrated circuit (6SIC), or one or more integrated circuits configured to implement the embodiments of this application. The smart device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more 6SICs.

[0179] Memory 606 is used to store program 610. Memory 606 may include high-speed ROM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0180] Specifically, program 610 can be used to cause processor 602 to perform the operations corresponding to the aforementioned data processing method.

[0181] The specific implementation of each step in program 610 can be found in the corresponding steps and units described in the above data processing method embodiments, and will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the devices and modules described above can be referred to the corresponding process descriptions in the foregoing method embodiments, and will not be repeated here.

[0182] It should be noted that, depending on the implementation needs, the various components / steps described in the embodiments of this application can be broken down into more components / steps, or two or more components / steps or parts of the operation of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of this application.

[0183] The methods described in the embodiments of this application can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, R6M, floppy disk, hard disk, or magneto-optical disk), or as computer code downloaded over a network that is originally stored in a remote recording medium or a non-transitory machine-readable medium and will be stored in a local recording medium. Thus, the methods described herein can be stored as software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as 6SiC or FPG6). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., R6M, ROM, flash memory, etc.) capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the data processing methods described herein. Furthermore, when a general-purpose computer accesses code used to implement the data processing methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for executing the data processing methods shown herein.

[0184] Those skilled in the art will recognize that the units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this application.

[0185] The above embodiments are only used to illustrate the embodiments of this application, and are not intended to limit the embodiments of this application. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of this application. Therefore, all equivalent technical solutions also fall within the scope of the embodiments of this application, and the patent protection scope of the embodiments of this application should be defined by the claims.

Claims

1. A data processing method for generating a description information profile associated with a target object, comprising: obtaining a description framework and a plurality of materials of a plurality of types associated with the target object, the description framework comprising attribute selection information corresponding to the target object; performing at least one processing on each of the materials to obtain attribute information of the material, the attribute information comprising an attribute level and an attribute content, wherein the attribute information is obtained by performing cleaning processing and standardization processing on an attribute sentence determined according to the plurality of materials; selecting target materials from the materials, the attribute content and the attribute level of which match the attribute selection information of the description framework; and generating the description information profile according to the description framework and the target materials, wherein the description information profile is generated according to a sorting result, the sorting result is obtained by sorting material paragraphs according to attribute levels corresponding to the material paragraphs formed according to a theme order indicated by the description framework and the target materials, and the material paragraphs are obtained by combining materials of which the attribute contents belong to a same theme. The plurality of types of materials comprise picture type materials and / or text type materials, and the text type materials comprise at least one of text obtained by performing speech recognition on audio type materials and original text. The performing at least one processing on each of the materials to obtain attribute information of the material comprises: performing image-text recognition on the picture type materials to obtain attribute sentences corresponding to the picture type materials; and / or performing attribute mining on the text type materials to obtain attribute sentences corresponding to the text type materials. The performing at least one processing on each of the materials to obtain attribute information of the material further comprises: performing cleaning processing on the attribute sentences according to semantic information of the attribute sentences; and performing standardization processing on attribute contents corresponding to the attribute sentences remaining after the cleaning processing to determine the attribute contents and corresponding attribute levels of the attribute sentences as attribute information of the corresponding materials according to a result of the standardization processing. The attribute selection information of the description framework is used to indicate a theme order, target attributes selected according to the themes, and attribute values selected according to the target attributes.

2. The method of claim 1, wherein, The attribute level of the material is used to indicate a theme corresponding to the material, and the attribute content of the material is used to indicate attributes and attribute values of the target object described by the material. The selecting target materials from the materials, the attribute content and the attribute level of which match the attribute selection information of the description framework, comprises: determining candidate materials corresponding to attribute levels of each of the themes according to a theme order indicated by the attribute selection information of the description framework; and determining target materials from the candidate materials, attribute contents of which match target attributes according to the target attributes of each of the themes and attribute values selected according to the target attributes. The determining target materials from the candidate materials, attribute contents of which match target attributes according to the target attributes of each of the themes and attribute values selected according to the target attributes, comprises: ​ 3. The method of claim 2, wherein, ​ ​ ​ 4. The method of claim 1, wherein, ​ 5. The method of claim 4, wherein, ​ 6. The method of claim 4 or 5, wherein, ​ ​ ​ 7. The method of claim 6, wherein, ​ According to the target attribute corresponding to each of the themes, select a candidate material whose attribute content matches the target attribute; According to the attribute value selected by the target attribute, select a candidate material containing the attribute value selected by the target attribute from the matched candidate materials as the target material.

8. The method of claim 1, wherein, The method further includes: Sort the materials with the same attribute value according to the first heat information, and remove the repeated materials with a semantic similarity satisfying a set value through a longest common subsequence algorithm; Group the materials corresponding to the same attribute content into a material group after removing the repeated materials; Combine the material groups with the same theme into a material paragraph.

9. The method of claim 4 or 5, wherein, The method further includes: Sort the material paragraphs according to the attribute hierarchy corresponding to the theme order indicated by the description framework, and generate the description information archive according to the sorting result.

10. A data processing method for generating a description information archive related to a target object, including: displaying a plurality of types of materials corresponding to the target object, a processing option corresponding to each of the materials, and a description framework editing option on a display interface; receiving a triggering operation on the processing option corresponding to at least one of the materials on the display interface, processing the material to obtain attribute information of the material, and displaying the attribute information, wherein the attribute information is obtained through cleaning and standardization processing of an attribute sentence determined according to a plurality of materials, and the attribute information includes an attribute hierarchy and an attribute content; receiving an editing operation on the description framework editing option, and obtaining the description framework according to the editing result, wherein the description framework includes attribute selection information corresponding to the target object; selecting a target material from the materials, wherein the attribute content and the attribute hierarchy of the target material match the attribute selection information of the description framework; generating a description information archive according to the description framework and the target material, wherein the description information archive is generated according to a sorting result, the sorting result is obtained by sorting material paragraphs according to an attribute hierarchy corresponding to a theme order indicated by the description framework, and the material paragraphs are obtained by grouping materials with the same theme.

11. The method of claim 10, wherein, The description framework editing option includes a theme adjustment option for adjusting the theme order in the description framework, an attribute adjustment option for adjusting a target attribute selected by a theme, and an attribute value adjustment option for adjusting an attribute value of the target attribute.

12. The method of claim 11, wherein, The method further includes: receiving an adjustment operation on at least one of the theme adjustment option, the attribute adjustment option, and the attribute value adjustment option; determining the attribute selection information of the description framework according to the adjustment result, including a theme order indicated by a broadcast instruction, a target attribute selected by the theme, and an attribute value of the target attribute.

13. A data processing apparatus for generating a description information profile associated with a target object, comprising: a first obtaining module configured to obtain a description framework and a plurality of materials of a plurality of types associated with the target object, the description framework comprising attribute selection information corresponding to the target object; a first processing module configured to perform at least one processing on each material to obtain attribute information of the material, the attribute information comprising an attribute level and an attribute content, wherein the attribute information is obtained by performing cleaning processing and standardization processing on an attribute sentence determined according to the plurality of materials; a first matching module configured to select, from the materials, target materials whose attribute contents and attribute levels match the attribute selection information of the description framework; a first generating module configured to generate a description information profile according to the description framework and the target materials, wherein the description information profile is generated according to a sorting result obtained by sorting material paragraphs according to attribute levels corresponding to the material paragraphs formed according to a theme order indicated by the description framework and the target materials, the material paragraphs being obtained by combining materials whose attribute contents belong to a same theme.

14. A data processing apparatus for generating a description information profile associated with a target object, comprising: a display module configured to display, on a display interface, materials of a plurality of types corresponding to the target object, processing options corresponding to the materials, and a description framework editing option; a second processing module configured to receive a triggering operation on a processing option corresponding to at least one material in the display interface, process the material to obtain attribute information of the material, and display the attribute information, the attribute information comprising an attribute level and an attribute content; a receiving module configured to receive an editing operation on the description framework editing option, and obtain the description framework according to an editing result, the description framework comprising attribute selection information corresponding to the target object, wherein the attribute information is obtained by performing cleaning processing and standardization processing on an attribute sentence determined according to a plurality of materials; a second matching module configured to select, from the materials, target materials whose attribute contents and attribute levels match the attribute selection information of the description framework; a second generating module configured to generate a description information profile according to the description framework and the target materials, wherein the description information profile is generated according to a sorting result obtained by sorting material paragraphs according to attribute levels corresponding to the material paragraphs formed according to a theme order indicated by the description framework and the target materials, the material paragraphs being obtained by combining materials whose attribute contents belong to a same theme.

15. An electronic device comprising: a processor, a memory, a communication interface, and a communication bus, the processor, the memory, and the communication interface being in communication with each other through the communication bus. The memory is configured to store at least one executable instruction, and the executable instruction is configured to enable the processor to perform operations corresponding to the data processing method according to any one of claims 1-9 or any one of claims 10-12. 16.A computer storage medium having stored thereon a computer program, the program being executed by a processor to implement the data processing method according to any one of claims 1-9 or any one of claims 10-12.

Images