Content screening method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of Internet, in particular to a content screening method and device, electronic equipment and a storage medium, and aims to improve the content screening accuracy. The method comprises the following steps: presenting a content display interface; the content display interface is used for displaying at least one to-be-viewed content; in response to an interaction operation triggered based on first target content in the at least one to-be-viewed content, presenting at least one piece of key feature information corresponding to a target object in the first target content in the content display interface, and updating a screening feature set corresponding to the target object based on the at least one piece of key feature information; the key feature information is used for screening contents matched with the target object; and in response to a matching operation triggered for a second target content in the at least one to-be-viewed content, presenting a corresponding matching result. According to the method, the screening feature set for the object individual is established, and other contents are screened based on the screening feature set, so that the screening accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to a content screening method, device, electronic device, and storage medium. Background Art

[0002] With the popularization of computers and the development of the Internet, people use the Internet more and more frequently, and computer networks have gradually become an indispensable tool in people's daily lives. People can filter information on the Internet to find the key information they are concerned about.

[0003] Taking the browser as an example, in related technologies, when an object filters information in the browser, a fixed conditional filter is provided for the object, and the page is updated in real time on the front end to provide timely feedback to the object. On the back end, data that meets the conditions is filtered based on key information. For example, in scenarios such as news and e-commerce, the object is provided with filtering based on conditions such as time, brand, and location. However, the object can only filter content in the browser within the limited set of key information provided by the fixed conditional filter. In this way, the filtering results are obtained based only on the limited set of key information, and the filtering accuracy is low.

[0004] Therefore, how to improve the accuracy of content screening is an urgent problem to be solved. Summary of the Invention

[0005] Embodiments of the present application provide a content screening method, apparatus, electronic device, and storage medium to improve the accuracy of content screening.

[0006] An embodiment of the present application provides a content screening method, including:

[0007] Presenting a content display interface; the content display interface is used to display at least one content to be viewed;

[0008] In response to an interactive operation triggered by a first target content in the at least one content to be viewed, presenting at least one key feature information corresponding to a target object in the first target content on the content display interface, and updating a screening feature set corresponding to the target object based on the at least one key feature information; the key feature information is used to screen content matching the target object;

[0009] In response to a matching operation triggered for a second target content in the at least one content to be viewed, a corresponding matching result is presented; the matching result is a result obtained by comparing the second target content with at least one key feature information in the screening feature set.

[0010] It should be emphasized that in the specific implementation of the present application, the relevant data involved in the object during content screening, such as the first target content and second target content selected by the object each time listed above, the data generated by the interactive operation triggered by the object, the data generated by the matching operation triggered by the object, the matching results, key feature information, the screening feature set, etc.

[0011] When the above embodiments of this application are applied to specific products or technologies, the subject's permission or consent must be obtained, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0012] Another content screening method provided by an embodiment of the present application includes:

[0013] After receiving an interaction request for a first target content, performing feature analysis on a target object in the first target content to determine at least one key feature information corresponding to the target object, wherein the key feature information is used to filter content matching the target object;

[0014] Feeding back the at least one key feature information to the client, so that the client presents the at least one key feature information on a content display interface in response to an interactive operation triggered by the first target content, and updates a screening feature set corresponding to the target object based on the at least one key feature information;

[0015] After receiving a matching request for the second target content, the second target content is compared with at least one key feature information in the screening feature set to obtain a matching result;

[0016] The matching result is sent to the client, so that the client presents the matching result after responding to the matching operation triggered for the second target content.

[0017] An embodiment of the present application provides a content screening device, comprising:

[0018] A presentation unit, configured to present a content presentation interface; the content presentation interface is configured to present at least one content to be viewed;

[0019] a first response unit configured to, in response to an interactive operation triggered by a first target content in the at least one content to be viewed, present at least one key feature information corresponding to a target object in the first target content on the content display interface, and update a screening feature set corresponding to the target object based on the at least one key feature information; the key feature information being used to screen content matching the target object;

[0020] The second response unit is used to present a corresponding matching result in response to a matching operation triggered for a second target content in the at least one content to be viewed; the matching result is: a result obtained by comparing the second target content with at least one key feature information in the screening feature set.

[0021] Optionally, the first response unit is specifically configured to:

[0022] In response to the interactive operation triggered by dragging a first target content in the at least one content to be viewed to a preset area in the content display interface;

[0023] In response to calling a filter control based on a first target content in the at least one content to be viewed, the interactive operation is triggered by the filter control;

[0024] In response to the interactive operation being triggered by a preset pinch gesture on a first target content in the at least one content to be viewed.

[0025] Optionally, the first response unit is specifically configured to:

[0026] In response to an interactive operation triggered on a first target content in the at least one content to be viewed, presenting a filter control on the content display interface;

[0027] In response to an adjustment operation triggered on the filter control, presenting a feature adding control on the content display interface;

[0028] In response to a feature adding operation triggered on the feature adding control, at least one key feature information corresponding to the target object in the first target content is presented.

[0029] Optionally, the first response unit is specifically configured to:

[0030] In response to an interactive operation triggered by voice input in conjunction with a first target content in the at least one content to be viewed, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

[0031] Optionally, the first response unit is specifically configured to:

[0032] In response to an interactive operation triggered based on a first target content in the at least one content to be viewed, presenting a filter control on the content display interface;

[0033] In response to a call-up operation triggered on the filter control, calling up a voice input function;

[0034] In response to a voice input operation triggered by the voice input function, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

[0035] Optionally, the content display interface further includes a filter control, and the first response unit is specifically configured to:

[0036] In response to a call-up operation triggered on the filter control, calling up a voice input function;

[0037] In response to a voice input operation triggered by the voice input function in combination with a first target content in the at least one content to be viewed, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

[0038] Optionally, if there are multiple target objects in the first target content, the first response unit is specifically configured to:

[0039] In response to an interactive operation triggered based on a first target content in the at least one content to be viewed, randomly presenting a filter control corresponding to the target object on the content display interface; or

[0040] In response to an interactive operation triggered based on a first target content in the at least one content to be viewed, a plurality of filter controls corresponding to each of the target objects are presented on the content display interface.

[0041] Optionally, after randomly presenting a filter control corresponding to the target object on the content display interface, and before presenting a feature addition control on the content display interface, the first response unit is further configured to:

[0042] In response to a control viewing operation triggered by a filter control randomly presented on the content display interface, presenting filter controls corresponding to other target objects among the multiple target objects on the content display interface;

[0043] In response to a switching operation triggered by the filter controls corresponding to the other target objects, determining a filter control after switching;

[0044] The first response unit is specifically configured to:

[0045] In response to the adjustment operation triggered on the switched filter control, a feature addition control is presented on the content display interface.

[0046] Optionally, after the content display interface presents a plurality of filter controls corresponding to the target objects, respectively, and before the first response unit presents a feature addition control in response to an adjustment operation triggered on the filter control, the first response unit is further configured to:

[0047] In response to a switching operation triggered by the filter controls corresponding to the plurality of target objects, determining a filter control after switching;

[0048] The first response unit is specifically configured to:

[0049] In response to the adjustment operation triggered on the switched filter control, a feature addition control is presented on the content display interface.

[0050] Optionally, the first response unit is specifically configured to:

[0051] In response to a selection operation triggered on first target key feature information in the at least one key feature information, the first target key feature information is added to a screening feature set corresponding to the target object.

[0052] Optionally, the first response unit is specifically configured to:

[0053] Within a preset time period after presenting the at least one key feature information, in response to a selection operation on the first target key feature information in the at least one key feature information, and determining that the first target key feature information is not deselected within the preset time period, the first target key feature information is added to the screening feature set corresponding to the target object.

[0054] Optionally, the content display interface further includes a filter control, and the first response unit is further configured to:

[0055] In response to an adjustment operation triggered on the filter control, presenting a feature deletion control on the content display interface;

[0056] In response to a triggering operation on the feature deletion control, presenting each key feature information included in the current screening feature set;

[0057] In response to a removal operation on the second target key feature information in the current key feature information, the second target key feature information is deleted from the screening feature set.

[0058] Optionally, the second response unit is further configured to:

[0059] In response to a result viewing operation triggered for the matching result, each key feature information included in the current feature screening set is presented, and the third target key feature information that matches the second target content among the current key feature information is highlighted.

[0060] Optionally, the content display interface further includes a filtering control; the matching result includes: a matching degree obtained by comparing the second target content with at least one key feature information in the filtering feature set;

[0061] Then the second response unit is specifically used to:

[0062] In response to a matching operation triggered by the second target content and the filter control, presenting a matching control on the content display interface;

[0063] In response to a triggering operation on the matching control, a matching degree corresponding to the second target content is presented.

[0064] Optionally, the second response unit is specifically configured to:

[0065] In response to a matching operation triggered by dragging the second target content to the filter control, presenting a matching control on the content display interface;

[0066] The second response unit is specifically configured to:

[0067] In response to a matching operation triggered by dragging the second target content from the filter control to the matching control, a matching degree corresponding to the second target content is presented.

[0068] Another content screening device provided in an embodiment of the present application includes:

[0069] a first receiving unit configured to, upon receiving an interaction request for a first target content, perform feature analysis on a target object in the first target content to determine at least one key feature information corresponding to the target object, wherein the key feature information is used to filter content matching the target object;

[0070] a first feedback unit, configured to feed back the at least one key feature information to the client, so that the client presents the at least one key feature information on a content display interface in response to an interactive operation triggered by the first target content, and updates a screening feature set corresponding to the target object based on the at least one key feature information;

[0071] a second receiving unit, configured to, upon receiving a matching request for a second target content, compare the second target content with at least one key feature information in the screening feature set to obtain a matching result;

[0072] The second feedback unit is configured to send the matching result to the client, so that the client presents the matching result after responding to the matching operation triggered for the second target content.

[0073] Optionally, the first receiving unit is specifically configured to:

[0074] identifying at least one target object contained in the first target content;

[0075] For each target object, perform the following operations:

[0076] Collecting hotly discussed information related to a target object from the Internet; and extracting first description information of the target object from the text information corresponding to the first target content;

[0077] At least one key feature information corresponding to the target object is determined based on the collected hot discussion information and the extracted first description information.

[0078] Optionally, the first receiving unit is specifically configured to:

[0079] collecting network data related to the one target object;

[0080] For each network data, perform the following operations:

[0081] Extracting a summary of text information corresponding to a piece of network data to generate a text summary corresponding to the piece of network data;

[0082] Segmenting the text summary into words;

[0083] For each word, determining the frequency of occurrence of the word in the text summary and the inverse document frequency of the word in network data related to the one target object;

[0084] Based on the occurrence frequency and inverse document frequency corresponding to each word, hotly discussed information related to the target object is extracted from the text summary.

[0085] Optionally, before the first receiving unit extracts a summary of the text information corresponding to a piece of network data to generate a text summary corresponding to the piece of network data, the first receiving unit is further configured to:

[0086] Performing sentiment analysis on the one network data to determine sentiment information related to the one target object in the one network data;

[0087] extracting comment information related to the target object from the network data;

[0088] The first receiving unit is specifically configured to:

[0089] The sentiment information and the comment information are combined to extract a summary of the text information corresponding to a piece of network data, and generate a text summary corresponding to the piece of network data.

[0090] Optionally, the matching result includes: a matching degree obtained by comparing the second target content with at least one key feature information in the screening feature set;

[0091] Then the second receiving unit is specifically configured to:

[0092] extracting second description information of the target object from the text information corresponding to the second target content;

[0093] For each piece of second description information, matching the second description information with at least one key feature information in the screening feature set;

[0094] The matching degree is determined according to the number of the successfully matched second description information and the total number of the at least one key feature information.

[0095] An embodiment of the present application provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of any one of the above-mentioned content screening methods.

[0096] An embodiment of the present application provides a computer-readable storage medium, which includes a computer program. When the computer program is run on an electronic device, the computer program is used to enable the electronic device to perform the steps of any one of the above-mentioned content screening methods.

[0097] An embodiment of the present application provides a computer program product, which includes a computer program stored in a computer-readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, causing the electronic device to perform the steps of any one of the above-mentioned content screening methods.

[0098] The beneficial effects of this application are as follows:

[0099] The embodiment of the present application provides a content screening method, device, electronic device and storage medium. In the embodiment of the present application, by generating key feature information corresponding to the target object in the first target content according to the interactive operation of the object based on the first target content during the process of the object browsing the content, and based on these key feature information, updating the screening feature set related to the target object for the object, the key feature information in the screening feature set can be used to screen the content matching the target object during the subsequent browsing process of the object, which can help the object gradually clarify the vague preferences into needs and quantify them, and help the object to effectively refine the key feature information without affecting or forcibly interrupting the reading experience of the object, and lay the foundation for the subsequent improvement of screening efficiency.

[0100] Furthermore, when the object sees the second target content of interest, a matching operation for the second target content can be triggered. Based on this, the second target content can be directly matched with the key feature information in the screening feature set. According to the matching results, the object can determine whether it is necessary to further browse the second target content for in-depth understanding. While improving the accuracy of content screening, it also improves the efficiency of content acquisition.

[0101] In summary, this application can establish a personal screening feature set for the object according to the needs of the object, complete the prior link of content screening, and then use the screening feature set to match other content to be viewed. Since the screening feature set is established according to the personal needs of the object, the screening accuracy is greatly improved.

[0102] Other features and advantages of the present application will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present application. The purposes and other advantages of the present application can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0104] Figure 1 A schematic diagram of an optional application scenario provided in an embodiment of the present application;

[0105] Figure 2 , which is a flowchart of an implementation of a content screening method provided in an embodiment of the present application;

[0106] Figure 3A A schematic diagram of content to be viewed provided in an embodiment of the present application;

[0107] Figure 3B A schematic diagram of another content to be viewed provided in an embodiment of the present application;

[0108] Figure 4A A schematic diagram of a first target content provided in an embodiment of the present application;

[0109] Figure 4B A schematic diagram of another first target content provided in an embodiment of the present application;

[0110] Figure 4C A schematic diagram of another first target content provided in an embodiment of the present application;

[0111] Figure 5A A schematic diagram of establishing a filter provided in an embodiment of the present application;

[0112] Figure 5B Another schematic diagram of establishing a filter provided in an embodiment of the present application;

[0113] Figure 6A A schematic diagram of an interactive operation provided in an embodiment of the present application;

[0114] Figure 6B Another interactive operation diagram provided in an embodiment of the present application;

[0115] Figure 6C A schematic diagram of another interactive operation provided in an embodiment of the present application;

[0116] Figure 7 A schematic diagram of key feature information provided in an embodiment of the present application;

[0117] Figure 8 A schematic diagram of newly added key feature information provided in an embodiment of the present application;

[0118] Figure 9 A schematic diagram of another key feature information provided in an embodiment of the present application;

[0119] Figure 10 A schematic diagram of a voice input provided in an embodiment of the present application;

[0120] Figure 11 A schematic diagram of deleting key feature information provided in an embodiment of the present application;

[0121] Figure 12 A schematic diagram of a filter provided in an embodiment of the present application;

[0122] Figure 13 A filter switching diagram provided in an embodiment of the present application;

[0123] Figure 14A schematic diagram of another method for establishing multiple filters provided in an embodiment of the present application;

[0124] Figure 15 Another filter switching diagram provided in an embodiment of the present application;

[0125] Figure 16 A schematic diagram of a matching result provided in an embodiment of the present application;

[0126] Figure 17 Another matching result diagram provided in an embodiment of the present application;

[0127] Figure 18 Flowchart of another content screening method provided in an embodiment of the present application;

[0128] Figure 19 A flowchart of a matching result provided in an embodiment of the present application;

[0129] Figure 20 A schematic diagram of the structure of a content screening device according to an embodiment of the present application;

[0130] Figure 21 Schematic diagram of the structure of another content screening device in an embodiment of the present application;

[0131] Figure 22 A schematic diagram of the hardware structure of an electronic device to which an embodiment of the present application is applied;

[0132] Figure 23 The present invention is a schematic diagram of the hardware structure of another electronic device to which the embodiments of the present application are applied. DETAILED DESCRIPTION

[0133] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of the technical solutions of this application, but not all of them. Based on the embodiments described in this application document, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the technical solutions of this application.

[0134] The following is an introduction to some concepts involved in the embodiments of this application.

[0135] Key feature information: refers to important data fields or attributes used for content screening. It can be numerical, textual or date-type data, or logical or Boolean data, etc., which is not specifically limited in this article. In the embodiment of the present application, when performing content screening, key feature information is usually related to the screening conditions. Taking the key feature information as textual data as an example, if the object wants to screen cars according to the model, then the car-related attribute fields such as "2.0T" and "7DCT" can be used as the key feature information of the car. If the object wants to screen sports watches according to the price, then the price field "within 1,000 yuan" can be used as the key feature information of the sports watch; and so on.

[0136] Screening feature set: refers to a set composed of at least one key feature information of the target object, which meets the individual needs of the subject for the target object. Subsequently, the entire network information can be screened according to the screening feature set to find content that meets the individual needs of the subject.

[0137] Content feature filter: Simply referred to as a filter, this is a device used to filter and select specific content. It can determine and classify the suitability of content based on predefined rules, keywords, themes, or other features. In an embodiment of the present application, based on the subject's interaction with certain content, a filter specific to the subject can be determined. The filter can be presented as a filter control on the content display interface, and the content of the filter can be a set of filtering features. Based on the filter's set of filtering features, the subject can match other content to achieve filtering.

[0138] The Bidirectional and AutoRegressive Transformers (BART) model is a sequence generation model based on the attention mechanism (Transformer) architecture. It combines the features of a bidirectional encoder and an autoregressive decoder and can be used for a variety of natural language processing tasks, such as text summarization, machine translation, and dialogue generation. The core concept of the BART model is to learn language representation and generation capabilities by leveraging large-scale unsupervised data through pre-training and fine-tuning.

[0139] Prior conditions: refers to the conditions for subjectively estimating the likelihood of an event based on past experience and knowledge. In the embodiments of this application, the prior conditions refer to the key feature information set by the subject in the filter based on their own preferences. These key feature information are the prerequisites for matching other content.

[0140] Hotly debated information refers to information that has been publicly released and has received widespread attention, sparked social repercussions, and sparked heated discussion and debate. It generally reflects content that is of general interest to the public and often sparks heated discussion online. In this embodiment of the present application, hotly debated information refers to information that is relevant to the target object and that has attracted widespread attention and can be collected online.

[0141] The embodiments of the present application relate to artificial intelligence (AI) and machine learning technologies, and are designed based on computer vision (CV), speech technology (Speech Technology), natural language processing (NLP), and machine learning (ML) in artificial intelligence.

[0142] Artificial Intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also studies the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0143] Artificial intelligence (AI) technology is a comprehensive discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0144] Key technologies in speech technology include automatic speech recognition (ASR), text-to-speech (TTS), and voiceprint recognition. Enabling computers to hear, see, speak, and feel is the future direction of human-computer interaction, with speech becoming one of the most promising methods of human-computer interaction. Large model technology is revolutionizing the development of speech technology. Pre-trained models based on the Transformer architecture, such as the language pre-trained model (WavLM) and the dedicated speech signal processing chip (UniSpeech), offer strong generalization and versatility, enabling them to effectively handle a wide range of speech processing tasks.

[0145] In an embodiment of the present application, automatic speech recognition technology can be used to perform speech recognition on the video content in the first target content and the second target content, and key feature information can be extracted therefrom. Automatic speech recognition technology can also be used to perform speech recognition on the speech input by the object, and key feature information can be extracted therefrom.

[0146] Natural language processing (NLP) is a key area of research in computer science and artificial intelligence. It studies the theories and methods that enable effective communication between humans and computers using natural language. Natural language processing (NLP) integrates linguistics, computer science, and mathematics. Therefore, research in this field involves natural language—the language we use in everyday life—and is closely linked to the study of linguistics. Natural language processing technologies typically include text processing, semantic understanding, machine translation, robotic question answering, and knowledge graphs.

[0147] Computer vision (CV) is the science of making machines "see." Specifically, it refers to the use of cameras and computers to replace the human eye in identifying, tracking, and measuring objects. Furthermore, it involves image processing, which transforms the computer's image into an image more suitable for human observation or transmission to instrumentation. As a scientific discipline, computer vision studies related theories and technologies, aiming to build artificial intelligence systems capable of extracting information from images or multidimensional data. Computer vision technologies typically include image processing, image recognition, image semantic understanding, image retrieval, optical character recognition (OCR), video processing, video semantic understanding, video content / action recognition, 3D object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping, and other technologies. Common biometric recognition technologies include facial recognition and fingerprint recognition.

[0148] Machine learning (ML) is a multidisciplinary field that encompasses probability theory, statistics, approximation theory, convex analysis, and algorithmic complexity theory. It specifically studies how computers can simulate or implement human learning behaviors to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is at the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications span all areas of AI. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and self-learning.

[0149] In an embodiment of the present application, the first target content and the second target content can be analyzed based on natural language processing technology, computer vision technology and machine learning technology, so as to extract key feature information based on the first target content, and after analyzing the second target content, the second target content can be compared with the key feature information to obtain a matching result.

[0150] With the research and advancement of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, autonomous driving, drones, digital twins, virtual humans, robots, AIGC, conversational interaction, smart medical care, smart customer service, game AI, etc. It is believed that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0151] In addition, the content screening method in the embodiment of the present application also involves database technology.

[0152] A database, in short, can be thought of as a digital filing cabinet—a place where electronic files are stored, where objects can add, query, update, and delete data within them. A "database" is a collection of data stored in a specific way, shared by multiple objects, with minimal redundancy, and independent of applications.

[0153] For example, various data generated in the embodiments of the present application (such as key feature information, screening feature sets, matching results, etc.) can be stored in a database for subsequent use, etc.

[0154] The following is a brief introduction to the design concept of the embodiment of this application:

[0155] With the popularization of computers and the development of the Internet, people use the Internet more and more frequently, and computer networks have gradually become an indispensable tool in people's daily lives. People can filter information on the Internet to find the key feature information they are concerned about.

[0156] Taking the browser as an example, in the related technology, a fixed condition filter can be used to combine the front-end interaction and the back-end data processing to realize the content screening function. Specifically, when the object filters the content in the browser, the browsing page is updated in real time on the front end to provide timely feedback to the object, and a fixed condition filter is provided for the object. The fixed condition filter will provide the object with limited and fixed key feature information for the object to filter the content in the browsing page according to the key feature information. At the back end, based on these key feature information, the data that meets the conditions is filtered in the browsing page. For example, the fixed condition filter can provide the object with key feature information such as time, brand, and location in scenarios such as news and e-commerce for filtering. Since the object can only filter the content in the browser within the limited set of key feature information provided by the fixed condition filter. In this way, the screening result is obtained based only on the limited set of key feature information, and the screening accuracy is low.

[0157] In related technologies, content filtering can also be performed using images. Based on the image provided by the user, content similar to the image, such as news content or e-commerce products, can be found. However, this method requires the user to provide images containing the content they require, which places significant restrictions on the filtering conditions and reduces the user experience.

[0158] In view of this, the embodiments of the present application provide a content screening method, device, electronic device and storage medium. In the embodiments of the present application, by generating key feature information corresponding to the target object in the first target content according to the interactive operation of the object based on the first target content during the process of the object browsing the content, and based on these key feature information, updating the screening feature set related to the target object for the object, the key feature information in the screening feature set can be used to screen the content matching the target object during the subsequent browsing process of the object, which can help the object gradually clarify the vague preferences into needs and quantify them, and help the object to effectively refine the key feature information without affecting or forcibly interrupting the reading experience of the object, and lay the foundation for the subsequent improvement of screening efficiency.

[0159] Furthermore, when the object sees the second target content of interest, a matching operation for the second target content can be triggered. Based on this, the second target content can be directly matched with the key feature information in the screening feature set. According to the matching results, the object can determine whether it is necessary to further browse the second target content for in-depth understanding. While improving the accuracy of content screening, it also improves the efficiency of content acquisition.

[0160] In summary, this application can establish a personal screening feature set for the object according to the needs of the object, complete the prior link of content screening, and then use the screening feature set to match other content to be viewed. Since the screening feature set is established according to the personal needs of the object, the screening accuracy is greatly improved.

[0161] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application can be combined with each other if there is no conflict.

[0162] like Figure 1 , which is a schematic diagram of an application scenario of an embodiment of the present application. The application scenario diagram includes two terminal devices 110 and a server 120.

[0163] In the embodiment of the present application, the terminal device 110 includes but is not limited to mobile phones, tablet computers, laptop computers, desktop computers, e-book readers, intelligent voice interaction devices, smart home appliances, car terminals and other devices; a content screening-related client can be installed on the terminal device, which can be software (such as a browser, news software, e-commerce software, etc.), or a web page, applet, etc. The server 120 is a background server corresponding to the software or web page, applet, etc., or a server specifically used for content screening, which is not specifically limited in this application. The server 120 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and basic cloud computing services such as big data and artificial intelligence platforms.

[0164] It should be noted that the content screening method in each embodiment of the present application can be executed by an electronic device, which may be a terminal device 110 or a server 120, that is, the method can be executed by the terminal device 110 or the server 120 alone, or can be executed jointly by the terminal device 110 and the server 120.

[0165] For example, when the terminal device 110 and the server 120 jointly execute, a content screening service-related client (such as a browser) is deployed on the terminal device 110. When the client responds to the interactive operation triggered by the object based on the first target content, the corresponding interactive request can be sent to the server 120 through the terminal device 110. Then, the server 120 performs feature analysis on the target object in the first target content in response to the interactive request of the first target content, determines at least one key feature information corresponding to the target object, and then sends the at least one key feature information to the terminal device 110, so that the terminal device 110 responds to the interactive operation triggered based on the first target content through the client, and presents at least one key feature information on the content display interface, and updates the screening feature set corresponding to the target object based on the at least one key feature information.

[0166] Moreover, when the client responds to the matching operation triggered for the second target content, it can send a corresponding matching request to the server 120 through the terminal device 110. The server 120 compares the second target content with at least one key feature information in the screening feature set in response to the matching request for the second target content, obtains a matching result, and sends the matching result to the terminal device 110 so that the terminal device 110 presents the matching result.

[0167] In an optional implementation, the terminal device 110 and the server 120 may communicate via a communication network.

[0168] In an optional implementation, the communication network is a wired network or a wireless network.

[0169] It should be noted that Figure 1 The figures are only examples. In fact, the number of terminal devices and servers is not limited and is not specifically limited in the embodiments of this application.

[0170] In an embodiment of the present application, when there are multiple servers, the multiple servers can be combined into a blockchain, and the servers are nodes on the blockchain; as in the content method disclosed in the embodiment of the present application, the target content and related data involved can be saved on the blockchain, for example, the first target content, the second target content, key feature information, the screening feature set, the matching results, etc.

[0171] In addition, the embodiments of the present application can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, assisted driving and other scenarios.

[0172] In addition, it can be understood that in the specific implementation of this application, when it involves object information and other related data (such as object interaction operations, object key feature information, etc.), when the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0173] The following describes the content screening method provided by the exemplary embodiment of the present application in combination with the application scenarios described above and with reference to the accompanying drawings. It should be noted that the above application scenarios are only shown to facilitate understanding of the spirit and principles of the present application, and the implementation of the present application is not limited in this respect.

[0174] See Figure 2 As shown in FIG, a flowchart of an implementation method of a content screening method provided in an embodiment of the present application is shown. Taking a client (such as a browser) as an example, the specific implementation process of the method is as follows S21 to S23:

[0175] S21: The client presents a content display interface.

[0176] The content display interface is used to display at least one content to be viewed. In the embodiment of the present application, the content display interface can be an information flow interface presented by a client such as a browser, e-commerce software, or news software, or it can be a content details interface presented by a client such as a browser, e-commerce software, or news software. This application does not make specific limitations on this.

[0177] In the embodiment of the present application, the content to be viewed may be multimedia content, i.e., a human-computer interactive information exchange and dissemination medium that combines two or more media, including text, pictures, sounds, videos, etc.

[0178] Specifically, multimedia content can be articles, news, information, videos, music, etc.

[0179] For example, the content to be viewed can be picture content, text content, etc. in the content display interface, or it can be video content, content combined with pictures and text, etc. This application does not make specific limitations on this.

[0180] See Figure 3A As shown, it is a schematic diagram of content to be viewed provided in an embodiment of the present application. Figure 3A The content to be viewed in the middle left picture includes text content, image content, and video content, and the content to be viewed in the right picture includes video content and text content.

[0181] See Figure 3B As shown, it is a schematic diagram of another content to be viewed provided in an embodiment of the present application. Figure 3B The content to be viewed in the middle left picture is pure text content, and the content to be viewed in the right picture is pure image content.

[0182] It should be noted that the above Figure 3A or Figure 3B This is just a simple example. In fact, any combination of one or more content forms can be the content to be viewed in this application, and they will not be listed one by one here.

[0183] S22: In response to an interactive operation triggered by a first target content in at least one content to be viewed, the client presents at least one key feature information corresponding to a target object in the first target content on a content display interface, and updates a screening feature set corresponding to the target object based on the at least one key feature information.

[0184] Among them, key feature information is used to filter content that matches the target object.

[0185] Specifically, the first target content can be a content set, such as a combination of pictures and text, video content, etc. in the content to be viewed, or it can be a single content, such as pure video frame content, pure text fragment content, pure picture content, etc. in the content to be viewed.

[0186] Among them, when the object selects the first target content, it can be the complete content in the content set or single content, such as the entire picture, the entire video, the entire video frame, etc., or it can be partial content in the content set or single content, such as a partial area of the picture, a partial area of the video frame, a video clip, a word or paragraph in the text content, and a combination of the above partial content, etc. This application does not make specific limitations on this.

[0187] See Figure 4A As shown, it is a schematic diagram of a first target content provided in an embodiment of the present application. Figure 4A The content to be viewed in the middle left picture includes text content, video content, and picture content. The first target content selected by the subject is the picture content (i.e., the car picture) and text content. The content to be viewed in the right picture includes video content and text content. The first target content selected by the subject is the video content (i.e., a video titled "Which Cars Can I Buy with a Budget of 200,000 Yuan").

[0188] See Figure 4B As shown, it is a schematic diagram of another first target content provided in an embodiment of the present application. Figure 4B The content to be viewed in the middle left picture is pure text content, and the first target content selected by the object is text content (i.e., better power performance). The content to be viewed in the right picture is pure picture content, and the first target content selected by the object is picture content (i.e., car picture).

[0189] See Figure 4C As shown, it is a schematic diagram of another first target content provided in an embodiment of the present application. Figure 4C The content to be viewed in the middle left picture is video content and text content, and the first target content selected by the object is a video frame in the video.

[0190] Specifically, when an object browses the first target content containing the target object it needs, it can trigger an interactive operation on the first target content, thereby presenting at least one key feature information corresponding to the above target object on the content display interface, and updating the filtering feature set containing the key feature information corresponding to the above target object, so that when subsequent objects browse to content matching the above target object, they can filter it through the key feature information in the filtering feature set.

[0191] It should be emphasized here that the above process needs to generate key feature information based on the interactive operations triggered by the object. In this case, it involves data generated by the interactive operations triggered by the object multiple times and key feature information generated multiple times. The collection, use and processing of this data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0192] Furthermore, the acquisition of this data requires the permission or consent of the target subject. Specifically, before generating key feature information, the target subject can be asked about the interactive operations that need to be obtained. After the target subject agrees, the above analysis process can be further performed. Alternatively, the target subject can be prompted to obtain the interactive operations that need to be obtained to generate key feature information.

[0193] In addition, it should be noted that when an object uses a filter for the first time, it must first trigger the following interaction to establish the filter; after the filter is established, the object can update the filter when it triggers the following interaction again.

[0194] Since the filter is presented as a filter control on the content display interface, after the filter is created, the filter control will be presented on the content display interface.

[0195] like Figure 5A As shown, it is a schematic diagram of establishing a filter provided by an embodiment of the present application. Figure 5A When browsing car-related content for a target, drag the video frame titled "2023 New SUV, Buy a Car in July and See Them, Stable and Comfortable, 2.0T+7DCT, and 8AT Transmission, Starting at Only 80,000" to the preset area S501 in the lower right corner of the content display interface to create a filter. After creating the filter, Figure 5A As shown in the content display page of S502, at this time, there is a filter control as shown in S503 in the lower right corner of the content display page.

[0196] like Figure 5B As shown, it is another schematic diagram of establishing a filter provided by an embodiment of the present application. Figure 5B When browsing car-related content for an object, by long pressing the video titled "Which cars can be bought with a budget of 200,000", the component function is called up, which includes the following: Figure 5B The filter is established as shown in S504, thereby establishing the filter. After the filter is established, as shown in Figure 5B As shown in the content display page of S505, at this time, there is a filter control as shown in S506 in the lower right corner of the content display page.

[0197] Optionally, the interactive operation triggered by the first target content in the at least one content to be viewed includes at least one of the following:

[0198] Interaction method 1: Drag the target content to trigger the interaction operation.

[0199] Specifically, the client responds to an interactive operation triggered by dragging a first target content in at least one content to be viewed to a preset area in the content display interface.

[0200] Among them, the preset area can be an area pre-set at any position in the content display interface, such as an area pre-set at the lower right corner of the content display interface, or an area pre-set at the lower left corner of the content display interface, etc. This application does not make specific limitations.

[0201] In the following example, a subject browses car-related content in a video scene. For example, the lower right corner of the content display interface is the preset area for creating filters. The subject can create a filter by dragging the first target content to this area. Figure 6A The interactive operation diagram shown.

[0202] like Figure 6A As shown, it is a schematic diagram of an interactive operation provided by an embodiment of the present application. Specifically, Figure 6A When the subject browses car-related content to be viewed, the interactive operation is triggered by dragging a video frame from a video titled "What car is good to buy around 200,000 yuan?", which is the first target content, to the preset area S601 in the lower right corner of the content display interface.

[0203] Interaction method 2: Call up the filter control to trigger the interactive operation.

[0204] Specifically, the client responds to the interactive operation triggered by the filter control after calling the filter control based on the first target content in the at least one content to be viewed.

[0205] Among them, the object can call up the filter control by any preset operation method, such as long pressing the first target content, double clicking the first target content, or calling up by voice command, etc., which are not specifically limited in this article. The following takes the case where the object calls up the filter control by long pressing the first target content as an example. For example, by long pressing the graphic content, pure image content, and video content to call up the filter control, a filter is established. 。 For details, please refer to Figure 6B The interactive operation diagram shown.

[0206] like Figure 6B As shown, it is another interactive operation diagram provided by the embodiment of the present application. Specifically, Figure 6B When the object browses the car-related content to be viewed, by long pressing the video titled "Which cars can be bought with a budget of 200,000", the video is the first target content, and the component function is called up. The component function includes the following: Figure 6B The filter control (ie, filter) shown in S602 triggers the interactive operation through the filter control.

[0207] Interaction method three: pinch gesture triggers interactive operations.

[0208] Specifically, the client responds to an interactive operation triggered by a preset pinch gesture on a first target content in at least one content to be viewed.

[0209] The preset pinch gesture may be specifically for pinching multiple different first target contents at the same time, or may be for pinching multiple content segments in the same first target content at the same time.

[0210] In addition, the multiple different first target contents kneaded simultaneously may be contents in the same multimedia format or in different multimedia formats; similarly, the multiple content segments kneaded simultaneously may be in the same multimedia format or in different multimedia formats.

[0211] For example, two first target contents that are both pictures can be pinched simultaneously, or two first target contents whose first target content is a picture and whose first target content is text can be pinched simultaneously, or two first target contents whose first target content is a video frame and whose first target content is text can be pinched simultaneously. This application does not make any specific restrictions on this.

[0212] Next, we use objects to pinch two viewed images simultaneously to trigger an interaction and create a filter.

[0213] like Figure 6C As shown, it is another interactive operation diagram provided by the embodiment of the present application. Specifically, Figure 6C When the subject browses automobile-related content to be viewed, the subject pinches two pictures in the first target content at the same time to trigger an interactive operation.

[0214] In the embodiments of the present application, the filter control can be presented on the content display interface by triggering the above-mentioned interactive operation, or it can be pre-set in the browser, news software, etc. through an object so that the filter control is always presented on the content filter interface. If the object is pre-set, the filter control can be presented on the content display interface without being triggered by an interactive operation. This application does not make specific restrictions on this.

[0215] In addition, a filter for the target object can be established through the object's voice command. For example, the object can issue a voice command "establish a filter for cars". The filter established in this way does not contain key feature information. When key feature information is subsequently added to the filter, the key feature information can be described through voice commands. Alternatively, the object can also establish a filter and add key feature information at one time. For example, the object can issue a voice command "establish a filter for cars, and the car needs to be brand A", then a filter for the target object of car is established and the key feature information "Brand A" is added; and so on. This application does not make specific restrictions on this.

[0216] In the above implementation, the present application can establish a filter through a variety of interactive operations, expand the usage of the object, provide an object usage experience, and facilitate subsequent object content screening according to the filter.

[0217] It should be emphasized that the acquisition of data generated by the above-mentioned interactive operations and voice commands is also subject to the permission or consent of the subjects, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0218] After the object initially establishes a filter for the target object, it can continue to browse the content to be viewed in the content display interface. During the continuous browsing process, as the amount of content browsed by the object increases, the object can continuously adjust its own needs for the target object. At the same time, the object can add or delete key feature information in the filter according to its continuously adjusted needs.

[0219] In an embodiment of the present application, an optional implementation method for adding key feature information to the filter is as follows:

[0220] The object can trigger an interactive operation on the first target content that interests him in the above manner, and the client presents a filter control on the content display interface in response to the interactive operation triggered by the object on the first target content in at least one content to be viewed; further, the object can adjust the filter according to other content that interests him in the subsequent browsing of content, and the client presents a feature addition control on the content display interface in response to the adjustment operation triggered by the object on the filter control; then, the object can trigger a feature addition operation on the feature filtering control according to other content that interests him, and the client presents at least one key feature information corresponding to the target object in the first target content in response to the feature addition operation triggered by the object on the feature addition control.

[0221] Among them, the filter is presented in the content display interface in the form of a filter control, and the filter control includes a sub-level control for supplementing the key feature information in the filter, which can be referred to as a feature addition control in this article.

[0222] In the embodiments of the present application, the adjustment operation may be performed by dragging the filter control, or by double-clicking the filter control, etc., which is not specifically limited in this application. The feature addition operation may be performed by dragging the first target content to the feature addition control, or by selecting the first target content and then double-clicking the feature addition control, etc., which is not specifically limited in this application.

[0223] like Figure 7 As shown, it is a schematic diagram of key feature information provided in an embodiment of the present application. Figure 7 Continued to use Figure 5A In S502, after the object creates a filter control S503, as shown in Figure 7 The adjustment operation triggered by the filter control in S701 is to drag the filter control, and the content filter interface is displayed as follows: Figure 7 In the feature adding control shown in S702, after dragging the first target content selected by the object to the feature adding control, as shown in FIG. Figure 7 As shown in S703 , key feature information S704 corresponding to the target object (car) in the first target content is presented on the content display interface: gearbox 8AT, 7DCT, 2.0T.

[0224] In the above implementation, the object adjusts the filter by adding new operations to the key feature information in the filter, so that the prior conditions contained in the filter better meet its own needs, which facilitates subsequent screening operations.

[0225] In an embodiment of the present application, when an object adds key feature information through a feature addition control in a filter, there may be problems such as the object not needing or being uninterested in the key feature information refined by the filter. In this case, the object can select the refined key feature information and only add the key feature information it needs to the filter.

[0226] In the embodiment of the present application, when adding key feature information to the filter, an optional implementation method is as follows:

[0227] The object can select the first target key feature information it needs, and the client responds to the selection operation triggered on the first target key feature information in the at least one key feature information, and adds the first target key feature information to the screening feature set corresponding to the target object.

[0228] Specifically, after the object extracts key feature information through the filter or subsequent self-input information, all the extracted key feature information will be presented in the content display interface. The object can select the key feature information they need or are interested in by selecting it and add it to the filter feature set, that is, the filter. For details, please refer to Figure 8 Schematic diagram of the newly added key feature information shown.

[0229] like Figure 8 As shown, it is a schematic diagram of newly added key feature information provided in an embodiment of the present application. Figure 8 For continued use Figure 5B Assuming that when the subject browses the car-related content to be viewed, after extracting key feature information by triggering interactive operations or autonomously inputting information, such as Figure 8 As shown, Figure 8The extracted key feature information S802 is presented in the content display interface S801, including gearboxes 8AT, 7DCT, and 2.0T. The subject is interested in the extracted key feature information of gearboxes 8AT and 2.0T and wants to add them to the filter. Therefore, Figure 8 In the content display interface shown in S803, as shown in S804, if the object selects the gearbox 8AT and 2.0T in the key feature information, these two key feature information can be added to the filter.

[0230] In addition, in the content display interface, objects can be swiped to view more key feature information. For details, please refer to Figure 9 Schematic diagram of key feature information shown.

[0231] like Figure 9 As shown, it is a schematic diagram of another key feature information provided in an embodiment of the present application. Figure 9 For continued use Figure 5B Assuming that when the subject browses the car-related content to be viewed, after extracting key feature information by triggering interactive operations or autonomously inputting information, such as Figure 9 As shown, Figure 9 The refined key feature information S902 is displayed in the content display interface S901, including gearbox 8AT, 7DCT and 2.0T. At this time, the object slides the key feature information in S902 counterclockwise to view more key features, such as Figure 9 More key feature information S904 is presented in the content display interface S903, including three major items and domestically produced.

[0232] When the object performs a selection operation on key feature information, within a preset time period after presenting at least one key feature information, in response to a selection operation on the first target key feature information in at least one key feature information, and determining that the first target key feature information is not deselected within the preset time period, the first target key feature information is added to the screening feature set corresponding to the target object.

[0233] Among them, the preset time length can be 2 seconds, 3 seconds, etc., and a reasonable preset time length can be set based on experience. This application does not make any specific restrictions on this.

[0234] Specifically, when the object adds key feature information to the filter through a selection operation, the object selects the key feature information it needs or is interested in and waits for a preset time. For example, after waiting for 2 seconds, it is assumed that the object has completed the operation of adding the key feature information. If the object does not need or is not interested in the extracted key feature information, it is also assumed that the object has completed the operation of adding the key feature information after staying for 2 seconds.

[0235] In an embodiment of the present application, the object can wait for a preset time after selecting the first target key feature information and then add it to the filter, or it can be added to the filter immediately after selecting the first target key feature information, etc. This application does not make any specific restrictions on this.

[0236] In the above implementation, the object can perform a selection operation on the key feature information that he is interested in or needs, and then add the key feature information to his own filter, thereby making the object's filter more in line with his needs and facilitating subsequent content filtering based on the filter.

[0237] In an embodiment of the present application, the key feature information can be obtained by refining the first target content selected by the object through a filter, or the key feature information can be extracted by the object inputting information independently.

[0238] If autonomous input information is completed through object voice input, the client can respond to the interactive operation triggered by voice input in combination with at least one first target content in the content to be viewed, and present at least one key feature information describing the target object in the first target content obtained by recognizing the input audio on the content display interface.

[0239] Specifically, when an object uses voice input to extract key feature information, if there is no filtering control on the content display interface, it is necessary to trigger the interactive operation first, present the filtering control, and then perform voice input.

[0240] In the embodiment of the present application, an optional implementation method for extracting key feature information through voice input is as follows:

[0241] The object can trigger an interactive operation on the first target content in at least one content to be viewed in the above-mentioned manner, and the client presents a filter control on the content display interface in response to the interactive operation triggered by the object based on the first target content in at least one content to be viewed; further, the object can trigger a call-up operation on the filter control to perform voice input through the voice input function, and the client calls up the voice input function in response to the call-up operation triggered by the object on the filter control; then, the object can perform voice input based on the voice input function, and the client presents on the content display interface at least one key feature information describing the target object in the first target content obtained by recognizing the input audio in response to the voice input operation triggered by the object based on the voice input function.

[0242] Specifically, when an object extracts key feature information through voice input, if the object is pre-set so that the filter control is always presented by default in the content display interface, there is no need to trigger an interactive operation, and voice input can be performed directly after the filter control.

[0243] In the embodiment of the present application, another optional implementation method for extracting key feature information through voice input is as follows:

[0244] The object may trigger a call-up operation on the filter control to perform voice input through the voice input function, and the client invokes the voice input function in response to the call-up operation triggered on the filter control;

[0245] The object can perform voice input based on the voice input function. In response to the voice input operation triggered by the voice input function in combination with the first target content in at least one content to be viewed, the client presents at least one key feature information describing the target object in the first target content obtained by recognizing the input audio on the content display interface.

[0246] In an embodiment of the present application, in combination with the first target content in at least one content to be viewed, it means that when the object browses to the first target content, if there is a need or interest in the first target content, the browsed first target content is described through voice, and then the key feature information of the target object in the corresponding first target content is presented through a filter.

[0247] like Figure 10 As shown, it is a schematic diagram of a voice input provided in an embodiment of the present application. Figure 10 When browsing car-related content for an object, use Figure 5A In S502, after the object creates a filter control by triggering an interactive operation S503, the filter control can be double-clicked to complete the triggering operation of the filter control, thereby waking up the voice input function. Figure 10 As shown in S1001, the icon of the filter control S1002 in S1001 changes, indicating that the voice input function has been aroused at this time. Then, the subject performs voice input operation in combination with the car-related content browsed by himself. Figure 10 The content display interface shown in S1003 presents the key feature information described for the car in S1003 obtained by recognizing the input audio, such as Figure 10 As shown in S1004, its key feature information is gearbox 8AT, 7DCT, 2.0T.

[0248] In the embodiment of the present application, when the subject extracts key feature information by autonomously inputting information, the above-mentioned voice input method or text input method, etc., may be used, and the present application does not make any specific restrictions on this.

[0249] In the above implementation, the subject extracts key feature information by autonomously inputting information, which broadens the way in which the subject establishes prior conditions and improves the user experience of the subject.

[0250] It should be emphasized that the acquisition of data such as the above-mentioned self-input information is also subject to the permission or consent of the subject, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0251] In embodiments of the present application, it is also possible for an object to delete key feature information from a filter. For example, if an object, after browsing content and adding key feature information, feels that some key feature information in the filter no longer meets their needs, they may need to delete the key feature information in the filter, removing any redundant key feature information that the object no longer needs or is not interested in.

[0252] In the embodiment of the present application, when deleting key feature information in the filter, an optional implementation method is as follows:

[0253] The object triggers an adjustment operation on the filter, and the client responds to the adjustment operation triggered on the filter control and presents a feature deletion control on the content display interface; then, the object triggers an operation on the filter control, and the client responds to the triggering operation on the feature deletion control and presents each key feature information contained in the current filter feature set; then, the object can remove the key feature information it does not need, and the client responds to the removal operation on the second target key feature information in the current key feature information and deletes the second target key feature information from the filter feature set.

[0254] Among them, the filter is presented in the content display interface as a filter control, and the filter control also includes a sub-level control for deleting key feature information in the filter, which can be referred to as a feature deletion control in this article.

[0255] In the embodiments of the present application, the adjustment operation may be dragging the filter control, double-clicking the filter control, etc., which is not specifically limited in the present application. The feature deletion operation may be dragging the first target content to the feature deletion control, or double-clicking the feature deletion control after selecting the first target content, etc., which is not specifically limited in the present application.

[0256] like Figure 11 As shown, it is a schematic diagram of deleting key feature information provided by an embodiment of the present application. Figure 11 Continued to use Figure 5A In S502, after the object creates a filter control S503, as shown in Figure 11 The adjustment operation triggered by S1101 for the filter control is to drag the filter control, and the content filter interface is displayed as follows Figure 11 In the feature deletion control shown in S1102, drag the first target content selected by the object to the feature addition control, as shown in FIG. Figure 11As shown in S1103, the content display interface presents key feature information S1104 corresponding to the target object (car) in the first target content: transmission 8AT, 7DCT, 2.0T. Since the object no longer requires the key feature information 7DCT, the object removes the key feature information 7DCT shown in S1106 in the filter, as shown in S1105 of the content display interface, completing the deletion operation of 7DCT.

[0257] In the above implementation, the object adjusts the filter by adding and deleting key feature information in the filter, so that the prior conditions contained in the filter better meet its own needs, which facilitates subsequent screening operations.

[0258] In addition, for the key feature information in the filter, the object can edit and adjust it by double-clicking or long pressing the key feature information to make the key feature information more in line with their needs.

[0259] It should be emphasized that the above-mentioned data acquisition for adding and deleting key feature information is also subject to the permission or consent of the subject, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0260] It should be noted that the above adjustment of the key feature information in the filter is based on a target object as an example. If the first target content contains multiple target objects, or the object has filtering requirements for multiple target objects in the first target content, or is interested in multiple target objects, etc., different filters can be established for multiple target objects.

[0261] Unlike one target object, when filters are simultaneously established / updated for multiple target objects in the first target content, multiple filters will be presented on the content display interface. For each filter, the above method can be used to add key feature information and delete key feature information.

[0262] In an embodiment of the present application, when multiple filters are established based on multiple target objects, an optional implementation method is as follows:

[0263] The object can trigger an interactive operation on a first target content containing multiple target objects. In response to the interactive operation triggered based on the first target content in at least one content to be viewed, the client randomly presents a filtering control corresponding to a target object on the content display interface.

[0264] like Figure 12 As shown, it is a schematic diagram of a filter provided in an embodiment of the present application. Figure 12In the example, the object has filtering requirements for the target objects of cars, commercial housing, and sports watches. At this time, three corresponding filtering controls are generated for each of the three target objects, and one of the three filtering controls is randomly presented in the content display interface, such as Figure 12 In S1201 , the filter controls corresponding to the target object "car" are randomly presented. To distinguish the filter controls, text may be marked in the filter controls to distinguish them. For example, the text marked in the filter control shown in S1201 is "car."

[0265] In this case, the filter corresponding to the target object car is selected by default. When adding or deleting key feature information of the filter later, the above-listed key feature information addition and deletion operations can be performed on the filter control corresponding to the target object car in the content display interface, such as Figure 10 The listed key feature information is added in different ways, Figure 11 For details of the key feature information deletion methods listed, please refer to the above embodiments, and the repeated parts will be omitted.

[0266] After an object has established multiple filters based on multiple target objects, in response to an adjustment operation triggered by a filter control, before presenting a feature addition control on the content display interface, if the object only has a filtering requirement for one of the target objects during a browsing process, and the latter object hopes to present only one filter control on the content display interface, the object can select and switch among multiple filters.

[0267] In an embodiment of the present application, an optional implementation method for switching multiple filter controls is:

[0268] The subject can perform a control-viewing operation on a filter control in the content display interface. In response to the control-viewing operation triggered by a filter control randomly presented on the content display interface, the client presents filter controls corresponding to each of the multiple target objects on the content display interface. Furthermore, the subject can perform a switching operation on other filter controls. In response to the switching operation triggered by the filter controls corresponding to the other target objects, the client determines the switched filter controls.

[0269] like Figure 13 As shown, it is a filter switching schematic diagram provided in an embodiment of the present application. Figure 13 For continued use Figure 12 Assuming that the object can trigger the control viewing operation of the filter control randomly presented on the content display interface by long pressing, double clicking, etc., for example, the object long presses Figure 12 The filter control corresponding to the target object of the car shown in S1201 triggers the control viewing operation. At this time, the filter control in the content display interface is as follows: Figure 13As shown in S1301, the filter controls corresponding to the other two target objects, commercial housing and sports watches, are presented. The subject selects the filter control corresponding to the target object commercial housing. Figure 13 As shown in S1302 , the content display interface will present the filter control corresponding to the target object of the commercial housing selected by the subject, thereby realizing the switching of the filter control.

[0270] After the filter control is switched, operations such as adding or deleting key feature information and matching the second target content can be performed on the switched filter control. For example, in response to an adjustment operation triggered by the switched filter control, a feature addition control is presented on the content display interface to add key feature information. For another example, in response to an adjustment operation triggered by the switched filter control, a feature deletion control is presented on the content display interface to delete key feature information, and so on.

[0271] In the above embodiment, the subject can establish filters for different target objects, which facilitates the subject to simultaneously filter multiple target objects that require filtering, thereby improving the user experience of the subject.

[0272] In the embodiment of the present application, when multiple filters are established based on multiple target objects, another optional implementation is as follows:

[0273] The object can trigger an interactive operation on a first target content containing multiple target objects. In response to the interactive operation triggered based on the first target content in at least one content to be viewed, filtering controls corresponding to each of the multiple target objects are presented on the content display interface.

[0274] like Figure 14 As shown, it is a schematic diagram of another method of establishing multiple filters provided in an embodiment of the present application. Figure 14 In the example, the object has filtering requirements for the target objects of cars, commercial housing and sports watches. At this time, three corresponding filtering controls are generated for these three target objects respectively, and in the content display interface, such as Figure 14 S1401 presents the filter controls corresponding to the target object of car, the filter controls corresponding to the target object of commercial housing, and the filter controls corresponding to the target object of sports watch. In order to distinguish the filter controls, text can be marked in the filter controls to distinguish them. For example, the three filter controls shown in S1401 are marked with the text car, commercial housing, and sports watch respectively.

[0275] In this case, in the content display interface, the three filter controls can perform the above-listed key feature information addition and deletion operations, and the second target content matching listed below, such as Figure 10 The listed key feature information is added in different ways, Figure 11The methods of deleting the listed key feature information, Figure 16 For details of the second target content matching, etc., please refer to the corresponding embodiments, and the repeated parts will be omitted.

[0276] In an embodiment of the present application, when determining the filter control after switching, an optional implementation method is as follows:

[0277] The object can perform a switching operation on multiple filter controls. The client determines the switched filter controls in response to the switching operation triggered by the filter controls corresponding to the multiple target objects.

[0278] Specifically, if multiple filters have been established, a fixed filter control with a fixed icon can be presented in the content display interface. The switching mode of the fixed filter control can be triggered by double-clicking, long pressing, etc. Based on the switching operation triggered by the filter controls corresponding to multiple target objects, the filter control after switching can be determined. For details, please refer to Figure 15 Diagram of the filter switch shown.

[0279] like Figure 15 As shown, it is another filter switching diagram provided by the embodiment of the present application. Figure 15 As shown in S1501, it is a fixed filter in the content display interface. Assume that the subject has established three filters for the target objects of cars, commercial housing, and sports watches. The subject hopes to display the filter control corresponding to the target object of commercial housing in the content display interface. The subject can trigger the control viewing operation of the fixed filter in the content display interface by long pressing, double clicking, etc. For example, the subject long presses Figure 15 The fixed filter shown in S1501 triggers the control viewing operation. At this time, the filter control in the content display interface is as follows: Figure 15 As shown in S1502, the filter controls corresponding to the other two target objects, commercial housing and sports watches, are presented. If the subject selects the filter control corresponding to the target object commercial housing, he can refer to Figure 13 As shown in S1302 , the content display interface will present the filter control corresponding to the target object of the commercial housing selected by the subject, thereby realizing the switching of the filter control.

[0280] After the filter control is switched, the key feature information and other operations can be performed on the switched filter control. For example, in response to an adjustment operation triggered on the switched filter control, a feature addition control is presented on the content display interface to add key feature information.

[0281] In the above implementation, by providing convenient interactive operations to obtain the subject's primary target content during the subject's browsing process, the subject's vague preferences are gradually clarified into specific needs, namely key feature information. Without affecting or forcibly interrupting the subject's browsing experience, the subject is helped to effectively accumulate information, facilitating subsequent matching and screening of other content. Furthermore, the subject can establish filters for different target objects, facilitating the simultaneous screening of multiple target objects with screening requirements, thereby improving the subject's user experience.

[0282] It should be emphasized that the above-mentioned data acquisition such as switching operations on multiple filters has also obtained the permission or consent of the subject, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0283] After adjusting the filter, the subject can use the filter to find content that meets their criteria.

[0284] S23: The client presents corresponding matching results in response to the matching operation triggered for the second target content in the at least one content to be viewed.

[0285] The matching result is a result obtained by comparing the second target content with at least one key feature information in the screening feature set. The matching result also includes a matching degree obtained by comparing the second target content with at least one key feature information in the screening feature set;

[0286] In an embodiment of the present application, the matching operation can be any preset interaction method, for example, dragging the second target content to the filter, or selecting the second target content and double-clicking the filter, etc. This application does not make specific limitations on this.

[0287] Specifically, according to the adjusted filter, the object can use the filter to compare and filter the content of interest at any time during the subsequent browsing process. For example, if the object is interested in the second target content, the object can drag the second target content to the filter to trigger the matching operation. At this time, the matching results of the second target content will be presented in the content display interface.

[0288] In an embodiment of the present application, an optional implementation method for determining the matching degree of the second target content through a filter is as follows:

[0289] The object can perform a matching operation on the second target content and the filter control, and the client presents the matching control on the content display interface in response to the matching operation triggered based on the second target content and the filter control; then, the object can perform a trigger operation on the filter control, and the client presents the matching degree corresponding to the second target content in response to the trigger operation on the matching control.

[0290] The filter is presented in the content display interface as a filter control, and the filter control also includes a sub-level control for matching the second target content, namely, the matching control in this article.

[0291] In the above embodiment, this application compares the second target content selected by the object with the filter pre-established by the object that meets its own needs, presents the matching degree of the second target content, improves the accuracy of content screening, and thus provides the object with a better usage experience.

[0292] It should be emphasized that the acquisition of data for the above-mentioned second target content, matching operations, etc. has also obtained the permission or consent of the subject, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0293] In an embodiment of the present application, an optional implementation method for determining the matching degree of the second target content through the sub-level control of the filter is as follows:

[0294] The object can drag the second target content to the filter, and the client presents the matching control on the content display interface in response to the matching operation triggered by dragging the second target content to the filter control; then, the object can drag the second target content to the matching control, and the client presents the matching degree corresponding to the second target content in response to the matching operation triggered by dragging the second target content from the filter control to the matching control.

[0295] like Figure 16 As shown, it is a schematic diagram of a matching result provided in an embodiment of the present application. Figure 16 After the filter is established for the target object car, the second target content of interest is browsed in the subsequent browsing process. Figure 16 As shown in S1601, the object drags S1601 to the filter in the lower right corner, and then the content display interface is displayed as follows Figure 16 The object continues to drag S1601 to the matching control shown in S1602, and the matching degree corresponding to the second target content is presented in the content display interface. As shown in Figure S1603, the matching degree corresponding to the second target content is 65%.

[0296] In the above embodiment, this application is based on the sub-level control of the filter. By comparing the second target content selected by the object with the filter pre-established by the object that meets its own needs, the matching degree of the second target content is presented, thereby improving the accuracy of content screening and providing a better usage experience for the object.

[0297] It should be emphasized that the acquisition of data such as the matching degree generated by the above-mentioned matching operations has also been obtained with the permission or consent of the subjects, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0298] In addition, the object can also be further viewed through the matching results, and which key feature information in the filter the second target content specifically matches and which key feature information does not match is presented in the content display interface.

[0299] In the embodiment of the present application, when matching the second target content with the key feature information in the filter, an optional implementation method is as follows:

[0300] The object can perform a result viewing operation on the matching results. In response to the result viewing operation triggered for the matching results, the client presents the various key feature information contained in the current feature screening set, and highlights the third target key feature information that matches the second target content among the current key feature information.

[0301] In the embodiment of the present application, the second target content matches the third target key feature, which means that the key feature information extracted by the filter based on the second target content is the same as the third target key feature information. Figure 17 The matching result diagram is shown.

[0302] like Figure 17 As shown, it is another matching result schematic diagram provided in an embodiment of the present application. Figure 17 For continued use Figure 16 Assume that the object can be clicked Figure 16 The matching degree shown in S1603 triggers the result viewing operation, and then the key feature information contained in the current filter is presented in the content display interface, namely 2.0T, domestic, and three major parts. The third target key feature information that matches the second target content with the key feature information in the filter is 2.0T and three major parts. Therefore, Figure 17 As shown in S1701, the two key feature information of 2.0T and three major components are highlighted.

[0303] In the above embodiment, this application compares the second target content selected by the object with the filter pre-established by the object and meeting its own needs, presents the matching degree of the second target content and the key feature information of the specific match, thereby improving the accuracy of content screening and providing the object with a better usage experience.

[0304] It should be emphasized that the acquisition of data for the above-mentioned result viewing operations, second target content, third target key feature information, etc. has also been obtained with the permission or consent of the subject, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0305] In the above-mentioned embodiment, this application provides convenient interactive operations to obtain the dimensions of the object's attention during the object's browsing of the content display interface, helping the object to gradually clarify vague preferences into needs, that is, a filter containing key feature information, which helps the object to effectively extract information without affecting or forcibly interrupting the object's browsing experience, and lays the foundation for the subsequent improvement of screening accuracy.

[0306] After establishing the filter, when the object sees the second target content of interest, the object can directly evaluate the matching degree through the established filter in advance through convenient interaction with the filter control, and then judge whether it is necessary to further read the second target content and whether to have a deeper understanding based on the evaluation results. This improves the accuracy of filtering while also improving the efficiency of content acquisition.

[0307] In summary, this application can establish a filter for individual subjects according to their needs, complete the prior link of content screening, and greatly improve the screening accuracy and efficiency.

[0308] In the case where multiple filters are established, a filter that is presented immediately, or a filter that is presented after object switching, can also perform a matching operation on the second target content, such as Figure 16 For details of the second target content matching, etc., please refer to the above embodiment, and the repeated parts will not be repeated. It should be noted that the above mainly illustrates the content screening method in the embodiment of the present application from the client side. The following further illustrates the content screening method in the embodiment of the present application from the server side:

[0309] See Figure 18 FIG. 1 is a flowchart of an implementation method of a content screening method in an embodiment of the present application. Taking the server as the execution subject as an example, the specific implementation process of the method is as follows S181 to S184:

[0310] S181: After receiving an interaction request for a first target content, the server performs feature analysis on a target object in the first target content to determine at least one key feature information corresponding to the target object. The key feature information is used to filter content matching the target object.

[0311] The interaction request is sent by the client in response to an interaction operation triggered by a first target content in at least one content to be viewed.

[0312] In an embodiment of the present application, the client first generates and sends a corresponding interaction request to the server in response to an interaction operation triggered by an object based on a first target content in at least one content to be viewed. The interaction request may include the object's identity information, identification information of the first target content, content details, etc. After receiving the interaction request, the server identifies the target object contained in the first target content, and then performs feature analysis on the target object in the first target content to extract key feature information.

[0313] Specifically, since the first target content can be a picture, text, video, video frame, voice, etc., when performing feature analysis on the target object in the first target content, at least one of the following technologies can be used: natural language processing technology, computer vision technology, and voice technology.

[0314] When performing feature analysis on the target object, the server will perform information cleaning and information extraction on the target object and its corresponding full-network information based on the first target content and the multimedia form of the full-network information collected based on the target object in the first target content, through natural language processing technology, computer vision technology, voice technology, etc., so as to obtain the true characteristics of the target object, that is, the key feature information.

[0315] In the embodiment of the present application, when performing feature analysis on a target object, an optional implementation method is as follows:

[0316] First, the server identifies at least one target object contained in the first target content.

[0317] Then, for each target object, the server performs the following operations:

[0318] The server collects hot information related to a target object from the network; and extracts first descriptive information of a target object from text information corresponding to the first target content; the server determines at least one key feature information corresponding to a target object based on the collected hot information and the extracted first descriptive information.

[0319] Hot topics refer to information collected online that is relevant to the target object and attracts widespread attention. For example, if the target object is a car, hot topics could include information about the safety and cost-effectiveness of a particular company's cars, or other users' user reviews of a particular car model, and so on.

[0320] The first description information refers to text information extracted from the first target content in the form of multimedia, such as video, text, picture, or a combination of picture and text.

[0321] If the multimedia format of the first target content is text, there is no need to extract text information from the target object, and the text can be directly used as the first description information.

[0322] If the multimedia form of the first target content is a combination of pictures and text, or a combination of video frames and text, or the multimedia form is pure picture content, the first target content needs to be converted into text information.

[0323] In the embodiments of the present application, text information in graphic content, picture content, video frames in a video, etc. can be extracted through OCR technology in computer vision technology, or graphic content, picture content, video frames in a video can be converted into text information through natural language processing technology, etc. This application does not make specific limitations on this.

[0324] Taking the conversion through OCR technology as an example, it is usually achieved using tools such as Tesseract engine and ABBYY FineReader. The specific steps include preprocessing, text detection, character recognition, post-processing, formatting and exporting. For each step, the specific operations are as follows:

[0325] Preprocessing: This step is used to improve the image quality of the image, including adjusting the brightness and contrast of the image, denoising the image, and binarizing the image to improve the accuracy of subsequent text recognition.

[0326] Text detection: This step uses text detection algorithms, such as the Canny edge detection algorithm and end-to-end text detection algorithms based on deep learning, to determine the location of text in the image.

[0327] Character recognition: This step recognizes each character in the image as the corresponding text. This can be achieved through traditional pattern matching technology, modern deep learning methods, etc. For example, Figure 4C In the example, the first target content selected by the object is a graphic content containing the text "Model I and Model J both have it", then through this step, the text "Model I and Model J both have it" in the first target content can be identified.

[0328] Post-processing: This step can correct errors in the above character recognition process, including grammar and context proofreading.

[0329] Formatting: This step organizes the recognized text into structured output according to the format and layout of the original document.

[0330] Export: This step saves the processed text as a file, such as a text (TXT) file or a document (DOC) file.

[0331] It should be noted that, in the above, the first description information is extracted based on the image content or the video frame content in the video selected for the object, and in the following, the first description information is extracted based on the complete video content selected for the object.

[0332] If the multimedia form of the first target content is pure video content, the first target content needs to be converted into text information.

[0333] In an embodiment of the present application, a video can be captured into multiple video frames, and then each video frame is converted using the above-mentioned method of converting an image into text, and the audio in the video is extracted, and then the audio is converted into text. Specifically, the steps include video processing, speech recognition, video frame conversion, audio extraction, and text processing. For each step, the specific operations are as follows:

[0334] Video processing: This step can capture video frames of the video through the Open Source Computer Vision Library (OpenCV), and can also transcode, crop, and extract video frames of the video through the Free Audio and Video Codec (ffmpeg); etc. This application does not make specific restrictions on this.

[0335] Video frame conversion: Taking the conversion using OCR technology as an example, this step can use tools such as the Tesseract engine and ABBYY FineReader to convert each video frame into text.

[0336] Audio extraction: This step extracts the audio from the video to facilitate subsequent speech recognition.

[0337] Speech recognition: This step can be implemented using ASR technology, TTS technology, NLP technology, etc. Taking ASR technology as an example, the speech interaction and text conversion technology (SpeechRecognition) and Google Cloud Speech-to-Text (Google Cloud Speech-to-Text) in ASR technology can be used.

[0338] Text processing: This step can use NLP technology to process and analyze the recognized text.

[0339] In addition, the object can also input the content that he needs or is interested in to the client by autonomously inputting information. The client then sends the content to the server to extract the first description information. The autonomous input information can be in the form of voice audio input by the object, or text information, picture information, etc. input by the object. This application does not make specific restrictions on this.

[0340] Specifically, taking the example of an object autonomously inputting information through voice input, if the object uses voice input to obtain key feature information in combination with the first target content, the client sends the voice input by the object to the server, and after the server receives the voice, it converts the voice into text.

[0341] When converting speech to text, taking ASR technology as an example, you can use the Google Cloud Speech-to-Text tool to perform speech-to-text conversion. The specific steps include data preparation, feature extraction, model training, model optimization, and deployment. For each step, the specific operations are as follows:

[0342] Data preparation: This step collects a large amount of audio data and corresponding text as training data.

[0343] Feature extraction: This step extracts features from the audio data. Specifically, it can be implemented using Mel Frequency Cepstral Coefficients (MFCCs), and the corresponding Mel Spectrogram is generated based on the MFCCs.

[0344] Model training: This step uses deep learning frameworks such as TensorFlow and PyTorch to train acoustic and language models.

[0345] Model optimization: This step improves the generalization capabilities of the acoustic model and language model through regularization, parameter adjustment, data augmentation, and other technologies.

[0346] Deployment: This step deploys the trained acoustic model and language model to the server, providing real-time, batch processing speech recognition services.

[0347] For example, if the object browses to Figure 8 In the content display interface shown in S801, the object is combined with the first target content, that is, the video title is "2023 new SUV, buy a car and see them in July, stable and comfortable, 2.0T+7DCT, also equipped with 8AT gearbox, starting at only 80,000 yuan", and the voice input "Car models include 2.0T+7DCT, also equipped with 8AT gearbox" is input. The client sends the voice to the server. After receiving the voice, the server converts the voice into the first description information "Car models include 2.0T+7DCT, also equipped with 8AT gearbox" through the above-mentioned voice-to-text method.

[0348] The above is a feature analysis of the first target content, and the first description information is extracted. The following is an introduction to the hotly discussed information collected from the entire network.

[0349] In an embodiment of the present application, when collecting hot discussion information related to a target object from the network, an optional implementation method is as follows:

[0350] First, the server collects network data related to the target object; then, for each piece of network data, the server performs the following operations:

[0351] The server extracts a summary of the text information corresponding to the network data and generates a text summary corresponding to the network data; the server divides the text summary into words; for each word, the server determines the frequency of occurrence of the word in the text summary and the inverse document frequency of the word in the network data related to a target object; finally, based on the frequency of occurrence and inverse document frequency corresponding to each word, the server extracts hotly discussed information related to a target object from the text summary.

[0352] The network data may include comments on the target object by other objects, discussions on the target object in social media, news reports on the target object, etc., and this application does not make any specific restrictions on this.

[0353] In an embodiment of the present application, when extracting hotly discussed information from network data, the network data is first preprocessed, and the text information corresponding to the preprocessed network data is processed into a text summary, and then the hotly discussed information is extracted from the text summary.

[0354] Among them, after collecting the network data, it is necessary to clean the network data, specifically including removing duplicates, missing values and noise to ensure the quality of the network data.

[0355] After cleaning the network data, you can perform text preprocessing on the network data, which includes steps such as word segmentation, stop word processing, stemming, and lemmatization. For each step, the specific operations are as follows:

[0356] Word segmentation: This step performs word segmentation on network data. For example, Chinese sentences can be split into short sentences and words, and foreign sentences can be split into words and phrases.

[0357] Stop word processing: This step is used to remove common stop words in the words obtained after splitting to reduce noise.

[0358] Stemming and lemmatization: This step is used to convert words into their original form to reduce the impact of word form changes on the analysis.

[0359] After the above-mentioned text preprocessing operation, the text information corresponding to the network data is first summarized and extracted. In the embodiment of the present application, the process uses the BART model, which specifically includes the steps of pre-training, fine-tuning and inference. For each step, the specific operations are as follows:

[0360] Pre-training: This step includes two parts: input representation and pre-training task.

[0361] The BART model uses a bidirectional and autoregressive approach to represent input. It learns a bidirectional representation of the text by masking some parts of the input text and then generates the rest of the text through autoregression.

[0362] During the pre-training phase, the BART model performs multiple self-supervised tasks, one of which is to mask a portion of the input text and let the model predict the content of the masked portion. Another task is to use autoregressive generation to maximize the log-likelihood of the entire text sequence.

[0363] Fine-tuning: This step includes two parts: task transformation and loss function.

[0364] Task conversion refers to the process of converting the BART model into a text summarization model during the fine-tuning step. In this case, the encoder-decoder structure of the BART model is used to generate a generalized text summary.

[0365] The loss function in the process of generating text summaries usually uses the log-likelihood of the generative task, which aims to minimize the difference between the generated text summary and the actual text summary.

[0366] Reasoning: This step refers to generating summaries. In the reasoning step, given the input text, that is, network data, the BART model can use its trained decoder to generate text summaries.

[0367] In an embodiment of the present application, a corresponding text summary can be generated based on the network data corresponding to a target object, and then hot information can be extracted based on the text summary.

[0368] Among them, the hot discussion information can be extracted from the text summary through the term frequency-inverse document frequency (TF-IDF) method, which includes word segmentation, term frequency (TF) calculation, inverse document frequency (IDF) calculation, TF-IDF value calculation, keyword extraction and other steps. For each step, the specific operations are as follows:

[0369] Word segmentation: Split the text summary into words or phrases to form a vocabulary.

[0370] Calculate term frequency (TF): For each word, calculate its frequency of occurrence in the text summary, i.e., term frequency. Term frequency refers to the number of times the word appears in the text summary divided by the total number of words in the document. You can refer to the following formula 1:

[0371]

[0372] Here, TF(t) refers to the word frequency of vocabulary t.

[0373] Calculate the inverse document frequency (IDF): For each word, calculate its inverse document frequency (IDF) in the entire corpus. The inverse document frequency is calculated by dividing the total number of documents by the number of documents containing the word, and then taking the logarithm of the quotient. You can refer to the following formula 2:

[0374]

[0375] Here, IDF(t) refers to the inverse document frequency of word t, +1 is to prevent the denominator from being zero, and the logarithm is taken to smooth the weight.

[0376] It should be noted that when determining the IDF(t) of word t, the entire corpus refers to all the network data searched when the target object in the first target content is searched throughout the entire network. For example, if a total of 50 documents related to cars are found when the target object is searched throughout the entire network, then the entire corpus is these 50 documents.

[0377] Calculate the TF-IDF value: This step multiplies the term frequency (TF) and the inverse document frequency (IDF) to obtain the TF-IDF value of each word, as shown in Formula 3 below:

[0378] TF-IDF(t)=TF(t)*IDF(t) (Formula 3)

[0379] Here, TF-IDF(t) refers to the TF-IDF value of word t, TF(t) refers to the term frequency of word t, and IDF(t) refers to the inverse document frequency of word t.

[0380] Keyword extraction: Sort the words in the text summary based on the calculated TF-IDF value. Since the higher the TF-IDF value, the more frequently the word appears in the text summary, you can sort from high to low according to the TF-IDF value and select the top-ranked words as the hot information.

[0381] In an embodiment of the present application, in order to make the second description information extracted from the network data related to the first target content selected by the object more accurate and detailed, a text summary can also be extracted based on the emotional information and comment information in the network data, and then hotly discussed information can be extracted.

[0382] When extracting text summaries based on sentiment and comment information contained in web data, an optional implementation is as follows:

[0383] Perform sentiment analysis on network data to determine sentiment information related to the target object in the network data; and / or extract comment information related to the target object from the network data.

[0384] Furthermore, the text information corresponding to the network data is abstracted and extracted in combination with the sentiment information and / or comment information to generate a text abstract corresponding to the network data.

[0385] Specifically, sentiment analysis technology can be used on network data to determine the sentiment polarity of comments about other objects in the network data and understand the emotional information of other objects towards the target object. For example, if the target object is car X, sentiment analysis technology can be used to determine whether the attitude held by other objects in their comments about car X is like or dislike.

[0386] You can also further extract comment information such as other objects' usage experience of the target object from the comments of other objects.

[0387] Combining the above sentiment information and comment information can generate a text summary corresponding to the network data.

[0388] After the server extracts the hotly discussed information and the first descriptive information, it can generate key feature information for the target object. For example, the hotly discussed information and the first descriptive information can be sorted together, sorted from high to low based on their relevance to the first target content selected by the object, and the information ranked at the top is presented as key feature information to the client for the object to view and use. Since the space on the content display interface in the client is limited, the number of key feature information generated can be pre-set. For example, the number of key feature information can be pre-set to 10. Then, the server can send these 10 key feature information to the client, so that the client can present them to the object and perform content screening.

[0389] In addition, to avoid generating duplicate key feature information, the server can compare the key feature information through a pre-trained Bidirectional Encoder Representations from Transformers (BERT) model. The BERT model can capture the meaning of key feature information in different contexts. By comparing the BERT embedding vectors of two key feature information, their semantic similarity can be determined.

[0390] Specifically, first install the Transformers library on the server, then load the pre-trained BERT model, encode the key feature information into the input format of the BERT model, obtain the embedding representation of the key feature information, and finally calculate the cosine similarity between the two embeddings. The closer the cosine similarity is to 1, the more similar the two embeddings are, and the closer it is to -1, the less similar the two embeddings are.

[0391] In the above implementation, the BART model in the field of deep learning is used to extract key feature information, and a filter that meets personal needs is established for the object, completing the prior link of content screening and facilitating subsequent matching and screening.

[0392] S182: The server feeds back at least one key feature information to the client, so that the client presents at least one key feature information on the content display interface in response to the interactive operation triggered based on the first target content, and updates the screening feature set corresponding to the target object based on the at least one key feature information.

[0393] After the server sends the key feature information to the client, the client can present the key feature information on the content display interface in response to the interactive operation of the object and the adjustment operation of the object based on the filter control.

[0394] Among them, the filter control is a device above other controls and elements in the content display interface.

[0395] For example, if the object is browsing the content display interface of the iPhone operating system, you can use the User Interface Window (UIWindow) to create a new window and set its windowLevel property. Then, use the present method in the User Interface ViewController (UIViewController) to present the filter above other controls and elements in the content display interface. If the object is browsing the content display interface of the Android operating system, you can use the System Alert Window permission to create a Window Manager instance to add a filter control and have it appear above other controls and elements in the content display interface.

[0396] S183: After receiving the matching request for the second target content, the server compares the second target content with at least one key feature information in the screening feature set to obtain a matching result.

[0397] Specifically, when the object is browsing other content, it browses to the second target content that it thinks may meet its needs. Therefore, the object drags the second target content to the filter and performs a matching operation on it. The client sends the matching request to the server, and the server can determine whether the second target content matches the key feature information predetermined by the object based on the filter.

[0398] Optionally, the matching result includes: when the second target content is compared with at least one key feature information in the screening feature set, the matching degree can be determined by:

[0399] First, second description information of a target object is extracted from the text information corresponding to the second target content; then, for each second description information, the second description information is matched with at least one key feature information in the screening feature set; finally, the matching degree is determined based on the number of successfully matched second description information and the total number of at least one key feature information.

[0400] Specifically, in response to the object's matching operation on the second target content, the client sends a matching request for the second target content to the server. The matching request may include the object's identity information, the second target content selected by the object, content details, etc. The server receives the matching request and extracts the second description information of the second target content using the above-mentioned method of extracting the first description information. Then, based on the BERT model, the server compares the second description information with the key feature information in the filter. The matching degree is determined based on the number of successfully matched second description information and the total number of key feature information in the filter. For details, please refer to the following formula 4:

[0401]

[0402] For example, if there are 10 key feature information in the filter, and 2 key feature information successfully matches the second description information, the matching degree is 20%.

[0403] In the above embodiment, the present application extracts the second description information of the second target content and matches the second description information with the key feature information in the filter predetermined by the object, thereby determining the matching degree between the second target content and the key feature information in the filter, making it easy for the object to determine whether the second target content meets its needs.

[0404] S184: The server sends the matching result to the client, so that the client presents the matching result after responding to the matching operation triggered for the second target content.

[0405] Optionally, when the matching result includes a matching degree, the object can view the key feature information that matches the second target content with the filter by double-clicking, long pressing, or other operations on the matching degree.

[0406] like Figure 19 As shown, it is a matching result flow chart provided by an embodiment of the present application. After browsing the content, the subject expresses doubts about whether the second target content is worthy of in-depth study, so it needs to match it with the prior conditions established by itself, that is, the filter. After the subject selects the second target content to be a priori, it drags it to the filter for matching, and the server extracts the second description information of the second target content. The second description information will be matched with the key feature information that the object has set in the filter. After the server determines the matching result of the second target content, it sends it to the client for easy viewing by the subject.

[0407] In the above-mentioned embodiment, the present application provides convenient interactive operations to obtain the dimensions of the object's attention during the object's browsing of the content display interface, helping the object to gradually clarify vague preferences into needs, that is, a filter containing key feature information, which helps the object to effectively extract information without affecting or forcibly interrupting the object's browsing experience, and lays the foundation for the subsequent improvement of screening accuracy.

[0408] After establishing the filter, when the subject sees the second target content of interest, he or she can directly evaluate the matching degree through the established filter in advance through convenient interaction with the filter control, and then judge whether to further read the second target content and whether to have a deeper understanding based on the evaluation results. This improves the accuracy of screening and the efficiency of content acquisition.

[0409] In summary, this application can establish a filter for individual objects according to the needs of the objects, complete the prior link of content screening, and use computer vision technology, voice technology, natural language processing technology, etc. in artificial intelligence to match the second target content with the key feature information in the filter, greatly improving the screening accuracy and efficiency.

[0410] Based on the same inventive concept, the embodiment of the present application also provides a content screening device. Figure 20 As shown, it is a schematic diagram of the structure of the content screening device 2000, which may include:

[0411] The presentation unit 2001 is configured to present a content presentation interface; the content presentation interface is configured to present at least one content to be viewed;

[0412] A first response unit 2002 is configured to, in response to an interactive operation triggered by a first target content in at least one content to be viewed, present at least one key feature information corresponding to a target object in the first target content on a content display interface, and update a screening feature set corresponding to the target object based on the at least one key feature information; the key feature information is used to screen content matching the target object;

[0413] The second response unit 2003 is used to present a corresponding matching result in response to a matching operation triggered for a second target content in at least one content to be viewed; the matching result is: a result obtained by comparing the second target content with at least one key feature information in the screening feature set.

[0414] Optionally, the first response unit 2002 is specifically configured to:

[0415] In response to an interactive operation triggered by dragging a first target content among the at least one content to be viewed to a preset area in the content display interface;

[0416] In response to an interactive operation triggered by the filter control after the filter control is called based on a first target content in the at least one content to be viewed;

[0417] In response to an interactive operation triggered by a preset pinch gesture on a first target content in the at least one content to be viewed.

[0418] Optionally, the first response unit 2002 is specifically configured to:

[0419] In response to an interactive operation triggered on a first target content in the at least one content to be viewed, presenting a filter control on the content display interface;

[0420] In response to an adjustment operation triggered on the filter control, presenting a feature addition control on the content display interface;

[0421] In response to a feature adding operation triggered on the feature adding control, at least one key feature information corresponding to the target object in the first target content is presented.

[0422] Optionally, the first response unit 2002 is specifically configured to:

[0423] In response to an interactive operation triggered by voice input in conjunction with a first target content in at least one content to be viewed, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on a content display interface.

[0424] Optionally, the first response unit 2002 is specifically configured to:

[0425] In response to an interactive operation triggered based on a first target content in at least one content to be viewed, presenting a filter control on the content display interface;

[0426] In response to a call-up operation triggered on the filter control, a voice input function is called up;

[0427] In response to a voice input operation triggered by a voice input function, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

[0428] Optionally, the content display interface further includes a filter control, and the first response unit 2002 is specifically configured to:

[0429] In response to a call-up operation triggered on the filter control, a voice input function is called up;

[0430] In response to a voice input operation triggered by a voice input function in combination with a first target content in at least one content to be viewed, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

[0431] Optionally, if there are multiple target objects in the first target content, the first response unit 2002 is specifically configured to:

[0432] In response to an interactive operation triggered by a first target content in at least one content to be viewed, randomly presenting a filter control corresponding to a target object on the content display interface; or

[0433] In response to an interactive operation triggered based on a first target content in at least one content to be viewed, filtering controls corresponding to each of a plurality of target objects are presented on a content display interface.

[0434] Optionally, after randomly presenting a filter control corresponding to a target object on the content display interface, the first response unit 2002 responds to an adjustment operation triggered on the filter control and before presenting a feature addition control on the content display interface, the first response unit 2002 is further configured to:

[0435] In response to a control viewing operation triggered by a filter control randomly presented on the content display interface, filter controls corresponding to other target objects among the multiple target objects are presented on the content display interface;

[0436] In response to a switching operation triggered by the filter controls corresponding to the other target objects, determining a filter control after switching;

[0437] The first response unit 2002 is specifically configured to:

[0438] In response to the adjustment operation triggered on the switched filter control, a feature addition control is presented on the content display interface.

[0439] Optionally, after the content display interface presents the filter controls corresponding to the multiple target objects, the first response unit 2002 responds to the adjustment operation triggered on the filter control and before the content display interface presents the feature addition control, the first response unit 2002 is further configured to:

[0440] In response to a switching operation triggered by the filter controls corresponding to the plurality of target objects, determining a filter control after switching;

[0441] The first response unit 2002 is specifically configured to:

[0442] In response to the adjustment operation triggered on the switched filter control, a feature addition control is presented on the content display interface.

[0443] Optionally, the first response unit 2002 is specifically configured to:

[0444] In response to a selection operation triggered on the first target key feature information in the at least one key feature information, the first target key feature information is added to the screening feature set corresponding to the target object.

[0445] Optionally, the first response unit 2002 is specifically configured to:

[0446] Within a preset time period after presenting at least one key feature information, in response to a selection operation on the first target key feature information in the at least one key feature information, and determining that the first target key feature information is not deselected within the preset time period, the first target key feature information is added to the screening feature set corresponding to the target object.

[0447] Optionally, the content display interface further includes a filter control, and the first response unit 2002 is further configured to:

[0448] In response to an adjustment operation triggered on the filter control, presenting a feature deletion control on the content display interface;

[0449] In response to a trigger operation on a feature deletion control, presenting each key feature information included in the current filtered feature set;

[0450] In response to a removal operation on the second target key feature information in each current key feature information, the second target key feature information is deleted from the screening feature set.

[0451] Optionally, the second response unit 2003 is further configured to:

[0452] In response to a result viewing operation triggered for a matching result, each key feature information included in the current feature screening set is presented, and the third target key feature information matching the second target content among the current key feature information is highlighted.

[0453] Optionally, the content display interface further includes a filtering control; the matching result includes: a matching degree obtained by comparing the second target content with at least one key feature information in the filtering feature set;

[0454] The second response unit 2003 is specifically used to:

[0455] In response to a matching operation triggered by the second target content and the filter control, presenting a matching control on the content display interface;

[0456] In response to a triggering operation on the matching control, a matching degree corresponding to the second target content is presented.

[0457] Optionally, the second response unit 2003 is specifically configured to:

[0458] In response to a matching operation triggered by dragging the second target content to the filter control, presenting a matching control on the content display interface;

[0459] The second response unit 2003 is specifically configured to:

[0460] In response to a matching operation triggered by dragging the second target content from the filter control to the matching control, a matching degree corresponding to the second target content is presented.

[0461] Based on the same inventive concept, the embodiment of the present application also provides another content screening device. Figure 21 As shown, it is a schematic diagram of the structure of the content screening device 2100, which may include:

[0462] The first receiving unit 2101 is configured to, upon receiving an interaction request for a first target content, perform feature analysis on a target object in the first target content to determine at least one key feature information corresponding to the target object, where the key feature information is used to filter content matching the target object;

[0463] The first feedback unit 2102 is configured to feed back at least one key feature information to the client, so that the client presents the at least one key feature information on the content display interface in response to the interactive operation triggered by the first target content, and updates the screening feature set corresponding to the target object based on the at least one key feature information;

[0464] The second receiving unit 2103 is configured to, after receiving a matching request for the second target content, compare the second target content with at least one key feature information in the screening feature set to obtain a matching result;

[0465] The second feedback unit 2104 is configured to send the matching result to the client, so that the client presents the matching result after responding to the matching operation triggered for the second target content.

[0466] Optionally, the first receiving unit 2101 is specifically configured to:

[0467] identifying at least one target object contained in the first target content;

[0468] For each target object, perform the following operations:

[0469] Collecting hotly discussed information related to a target object from the Internet; and extracting first description information of the target object from text information corresponding to the first target content;

[0470] At least one key feature information corresponding to a target object is determined based on the collected hot discussion information and the extracted first description information.

[0471] Optionally, the first receiving unit 2101 is specifically configured to:

[0472] Collect network data related to a target object;

[0473] For each network data, perform the following operations:

[0474] Extracting a summary of the text information corresponding to a network data to generate a text summary corresponding to the network data;

[0475] Split the text summary into words;

[0476] For each word, determine the frequency of occurrence of the word in the text summary and the inverse document frequency of the word in the network data related to a target object;

[0477] Based on the occurrence frequency and inverse document frequency of each word, hot discussion information related to a target object is extracted from the text summary.

[0478] Optionally, before the first receiving unit 2101 extracts a summary of the text information corresponding to a piece of network data and generates a text summary corresponding to the network data, the first receiving unit 2101 is further configured to:

[0479] Perform sentiment analysis on a network data to determine the sentiment information related to a target object in the network data;

[0480] Extract review information related to a target object from a network data;

[0481] The first receiving unit 2101 is specifically configured to:

[0482] Combining sentiment information and comment information, the text information corresponding to a network data is abstracted and extracted to generate a text summary corresponding to the network data.

[0483] Optionally, the matching result includes: a matching degree obtained by comparing the second target content with at least one key feature information in the screening feature set;

[0484] The second receiving unit 2103 is specifically configured to:

[0485] extracting second description information of a target object from the text information corresponding to the second target content;

[0486] For each piece of second description information, matching the second description information with at least one key feature information in the screening feature set;

[0487] The matching degree is determined according to the number of successfully matched second description information and the total number of the at least one key feature information.

[0488] For the convenience of description, the above parts are divided into modules (or units) according to their functions and described separately. Of course, when implementing this application, the functions of each module (or unit) can be implemented in the same or multiple software or hardware.

[0489] After introducing the content screening method and apparatus according to an exemplary embodiment of the present application, an electronic device according to another exemplary embodiment of the present application will be introduced next.

[0490] Those skilled in the art will appreciate that various aspects of the present application can be implemented as systems, methods, or program products. Therefore, various aspects of the present application can be specifically implemented in the following forms: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation that combines hardware and software aspects, which may be collectively referred to herein as a "circuit," "module," or "system."

[0491] Based on the same inventive concept as the above method embodiment, an electronic device is also provided in the embodiment of the present application. In one embodiment, the electronic device may be a server, such as Figure 1 In this embodiment, the structure of the electronic device can be as follows: Figure 22 As shown, it includes a memory 2201, a communication module 2203 and one or more processors 2202.

[0492] Memory 2201 is used to store computer programs executed by processor 2202. Memory 2201 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and programs required for running instant messaging functions, while the data storage area may store various instant messaging messages and operating instruction sets.

[0493] Memory 2201 may be a volatile memory, such as random-access memory (RAM); a non-volatile memory, such as read-only memory, flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); or any other medium capable of carrying or storing a desired computer program in the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 2201 may be a combination of the aforementioned memories.

[0494] The processor 2202 may include one or more central processing units (CPUs) or digital processing units, etc. The processor 2202 is configured to implement the above-mentioned method for generating recommended search information when calling the computer program stored in the memory 2201 .

[0495] The communication module 2203 is used to communicate with terminal devices and other servers.

[0496] The specific connection medium between the memory 2201, the communication module 2203 and the processor 2202 is not limited in the embodiment of the present application. Figure 22 The memory 2201 and the processor 2202 are connected via a bus 2204. Figure 22 The connections between the other components are shown in bold lines, which are only for illustration and are not intended to be limiting. The bus 2204 can be divided into an address bus, a data bus, a control bus, etc. For ease of description, Figure 22 The diagram shows a single thick line, but this does not indicate that there is only one bus or one type of bus.

[0497] The memory 2201 stores a computer storage medium, which stores computer executable instructions. The computer executable instructions are used to implement the content screening method of the embodiment of the present application. The processor 2202 is used to execute the above-mentioned content screening method, such as Figure 2 shown.

[0498] In another embodiment, the electronic device may also be other electronic devices, such as Figure 1 The terminal device 110 shown in FIG. In this embodiment, the structure of the electronic device can be as follows: Figure 23 As shown, it includes: a communication component 2310, a memory 2320, a display unit 2330, a camera 2340, a sensor 2350, an audio circuit 2360, a Bluetooth module 2370, a processor 2380 and other components.

[0499] The communication component 2310 is used to communicate with the server. In some embodiments, it may include a wireless fidelity (WiFi) module. The WiFi module is a short-range wireless transmission technology. Electronic devices can help users send and receive information through the WiFi module.

[0500] The memory 2320 can be used to store software programs and data. The processor 2380 executes various functions and data processing of the terminal device 110 by running the software programs or data stored in the memory 2320. The memory 2320 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. The memory 2320 stores an operating system that enables the terminal device 110 to run. In the present application, the memory 2320 can store the operating system and various application programs, and may also store a computer program that executes the content screening method of the embodiment of the present application.

[0501] The display unit 2330 can also be used to display information input by the user or information provided to the user, as well as a graphical user interface (GUI) of various menus of the terminal device 110. Specifically, the display unit 2330 may include a display screen 2332 provided on the front of the terminal device 110. The display screen 2332 may be configured in the form of a liquid crystal display, a light-emitting diode, etc. The display unit 2330 can be used to display the content display interface, filter controls, etc. in the embodiments of the present application.

[0502] The display unit 2330 can also be used to receive input digital or character information and generate signal input related to user settings and function control of the terminal device 110. Specifically, the display unit 2330 may include a touch screen 2331 arranged on the front of the terminal device 110, which can collect user touch operations on or near it, such as clicking a button, dragging a scroll box, etc.

[0503] The touch screen 2331 can be covered on the display screen 2332, or the touch screen 2331 and the display screen 2332 can be integrated to realize the input and output functions of the terminal device 110. The integrated display screen can be simply called a touch screen. In this application, the display unit 2330 can display applications and corresponding operation steps.

[0504] Camera 2340 can be used to capture still images, and users can publish images captured by camera 2340 through an application. Camera 2340 can be one or more. The object generates an optical image through the lens and projects it onto the photosensitive element. The photosensitive element can be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the optical signal into an electrical signal, which is then transmitted to processor 2380 for conversion into a digital image signal.

[0505] The terminal device may further include at least one sensor 2350, such as an acceleration sensor 2351, a distance sensor 2352, a fingerprint sensor 2353, and a temperature sensor 2354. The terminal device may also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, a light sensor, and a motion sensor.

[0506] The audio circuit 2360, speaker 2361, and microphone 2362 provide an audio interface between the user and the terminal device 110. The audio circuit 2360 can convert the received audio data into an electrical signal and transmit it to the speaker 2361, which converts it into a sound signal for output. The terminal device 110 may also be equipped with a volume button for adjusting the volume of the sound signal. On the other hand, the microphone 2362 converts the collected sound signal into an electrical signal, which is received by the audio circuit 2360 and converted into audio data. The audio data is then output to the communication component 2310 for transmission to, for example, another terminal device 110, or the audio data is output to the memory 2320 for further processing.

[0507] The Bluetooth module 2370 is used to exchange information with other Bluetooth devices having a Bluetooth module through the Bluetooth protocol. For example, the terminal device can establish a Bluetooth connection with a wearable electronic device (such as a smart watch) that also has a Bluetooth module through the Bluetooth module 2370 to exchange data.

[0508] The processor 2380 is the control center of the terminal device. It uses various interfaces and lines to connect various parts of the entire terminal. By running or executing software programs stored in the memory 2320 and calling data stored in the memory 2320, it performs various functions of the terminal device and processes data. In some embodiments, the processor 2380 may include one or more processing units; the processor 2380 may also integrate an application processor and a baseband processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the baseband processor mainly processes wireless communications. It is understandable that the above-mentioned baseband processor may not be integrated into the processor 2380. In the present application, the processor 2380 can run the operating system, application programs, user interface display and touch response, as well as the content screening method of the embodiment of the present application. In addition, the processor 2380 is coupled to the display unit 2330.

[0509] In some possible implementations, various aspects of the content screening method provided by the present application may also be implemented in the form of a program product, which includes a computer program. When the program product is run on an electronic device, the computer program is used to enable the electronic device to perform the steps of the content screening method according to various exemplary embodiments of the present application described above in this specification. For example, the electronic device may perform the following steps: Figure 2 Follow the steps shown in .

[0510] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0511] The program product of the embodiment of the present application may be a portable compact disc read-only memory (CD-ROM) and include a computer program, and can be run on an electronic device. However, the program product of the present application is not limited thereto. In this document, a readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with a command execution system, apparatus, or device.

[0512] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a readable computer program. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with a command execution system, apparatus, or device.

[0513] The computer program embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0514] The computer program for performing the operations of the present application can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as "C" or similar programming languages. The computer program can be executed entirely on the user electronic device, partially on the user electronic device, as a separate software package, partially on the user electronic device and partially on a remote electronic device, or entirely on a remote electronic device or server. In cases involving remote electronic devices, the remote electronic device can be connected to the user electronic device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external electronic device (for example, using an Internet service provider to connect through the Internet).

[0515] It should be noted that although several units or subunits of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, depending on the embodiment of the application, the features and functions of two or more units described above can be embodied in a single unit. Conversely, the features and functions of a single unit described above can be further divided and embodied by multiple units.

[0516] Furthermore, although the operations of the method of the present application are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0517] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0518] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0519] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0520] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0521] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A content screening method, characterized in that: The method comprises: Presenting a content display interface; the content display interface is used to display at least one content to be viewed; In response to an interactive operation triggered by a first target content in the at least one content to be viewed, presenting at least one key feature information corresponding to a target object in the first target content on the content display interface, and updating a screening feature set corresponding to the target object based on the at least one key feature information; the key feature information is used to screen content matching the target object; In response to a matching operation triggered for a second target content in the at least one content to be viewed, a corresponding matching result is presented; the matching result is a result obtained by comparing the second target content with at least one key feature information in the screening feature set.

2. The method according to claim 1, wherein The interactive operation triggered in response to the first target content in the at least one content to be viewed includes at least one of the following: In response to the interactive operation triggered by dragging a first target content in the at least one content to be viewed to a preset area in the content display interface; In response to calling a filter control based on a first target content in the at least one content to be viewed, the interactive operation is triggered by the filter control; In response to the interactive operation being triggered by a preset pinch gesture on a first target content in the at least one content to be viewed.

3. The method according to claim 1, wherein The step of presenting, in response to an interactive operation triggered by a first target content in the at least one content to be viewed, at least one key feature information corresponding to a target object in the first target content on the content display interface includes: In response to an interactive operation triggered on a first target content in the at least one content to be viewed, presenting a filter control on the content display interface; In response to an adjustment operation triggered on the filter control, presenting a feature adding control on the content display interface; In response to a feature adding operation triggered on the feature adding control, at least one key feature information corresponding to the target object in the first target content is presented.

4. The method according to claim 1, wherein The step of presenting, in response to an interactive operation triggered by a first target content in the at least one content to be viewed, at least one key feature information corresponding to a target object in the first target content on the content display interface includes: In response to an interactive operation triggered by voice input in conjunction with a first target content in the at least one content to be viewed, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

5. The method according to claim 4, wherein In response to the interactive operation triggered by voice input in conjunction with a first target content in the at least one content to be viewed, presenting, on the content display interface, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio, including: In response to an interactive operation triggered based on a first target content in the at least one content to be viewed, presenting a filter control on the content display interface; In response to a call-up operation triggered on the filter control, calling up a voice input function; In response to a voice input operation triggered by the voice input function, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

6. The method according to claim 4, wherein The content display interface further includes a filter control; the interactive operation triggered by voice input in conjunction with a first target content in the at least one content to be viewed, presenting at least one key feature information describing a target object in the first target content obtained by recognizing the input audio on the content display interface, including: In response to a call-up operation triggered on the filter control, calling up a voice input function; In response to a voice input operation triggered by the voice input function in combination with a first target content in the at least one content to be viewed, at least one key feature information describing a target object in the first target content obtained by recognizing the input audio is presented on the content display interface.

7. The method according to claim 3, wherein If there are multiple target objects in the first target content, presenting a filter control on the content display interface in response to the interactive operation triggered based on the first target content in the at least one content to be viewed includes: In response to an interactive operation triggered based on a first target content in the at least one content to be viewed, randomly presenting a filter control corresponding to the target object on the content display interface; or In response to an interactive operation triggered based on a first target content in the at least one content to be viewed, a plurality of filter controls corresponding to each of the target objects are presented on the content display interface.

8. The method according to claim 7, wherein After randomly presenting a filter control corresponding to the target object on the content display interface, in response to an adjustment operation triggered on the filter control, and before presenting a feature addition control on the content display interface, the method further includes: In response to a control viewing operation triggered by a filter control randomly presented on the content display interface, presenting filter controls corresponding to other target objects among the multiple target objects on the content display interface; In response to a switching operation triggered by the filter controls corresponding to the other target objects, determining a filter control after switching; The presenting of a feature adding control on the content display interface in response to the adjustment operation triggered on the filter control includes: In response to the adjustment operation triggered on the switched filter control, a feature addition control is presented on the content display interface.

9. The method according to claim 7, wherein After presenting a plurality of filter controls corresponding to the target objects on the content display interface, and in response to an adjustment operation triggered on the filter controls, and before presenting a feature addition control on the content display interface, the method further includes: In response to a switching operation triggered by the filter controls corresponding to the plurality of target objects, determining a filter control after switching; The presenting of a feature adding control on the content display interface in response to the adjustment operation triggered on the filter control includes: In response to the adjustment operation triggered on the switched filter control, a feature addition control is presented on the content display interface.

10. The method according to any one of claims 1 to 9, wherein: The updating of the screening feature set corresponding to the target object based on the at least one key feature information includes: In response to a selection operation triggered on first target key feature information in the at least one key feature information, the first target key feature information is added to a screening feature set corresponding to the target object.

11. The method according to claim 10, wherein In response to the selection operation on the first target key feature information in the at least one key feature information, adding the first target key feature information to the screening feature set corresponding to the target object includes: Within a preset time period after presenting the at least one key feature information, in response to a selection operation on the first target key feature information in the at least one key feature information, and determining that the first target key feature information is not deselected within the preset time period, the first target key feature information is added to the screening feature set corresponding to the target object.

12. The method according to any one of claims 1 to 9, wherein: The content display interface further includes a filter control, and the method further includes: In response to an adjustment operation triggered on the filter control, presenting a feature deletion control on the content display interface; In response to a triggering operation on the feature deletion control, presenting each key feature information included in the current screening feature set; In response to a removal operation on the second target key feature information in the current key feature information, the second target key feature information is deleted from the screening feature set.

13. The method according to any one of claims 1 to 9, wherein: The method further comprises: In response to a result viewing operation triggered for the matching result, each key feature information included in the current feature screening set is presented, and the third target key feature information that matches the second target content among the current key feature information is highlighted.

14. The method according to any one of claims 1 to 9, wherein: The content display interface further includes a filtering control; the matching result includes: a matching degree obtained by comparing the second target content with at least one key feature information in the filtering feature set; Then, in response to the matching operation triggered for the second target content in the at least one content to be viewed, presenting the corresponding matching result includes: In response to a matching operation triggered by the second target content and the filter control, presenting a matching control on the content display interface; In response to a triggering operation on the matching control, a matching degree corresponding to the second target content is presented.

15. The method according to claim 14, wherein The presenting of a matching control on the content display interface in response to a matching operation triggered by the second target content and the filter control includes: In response to a matching operation triggered by dragging the second target content to the filter control, presenting a matching control on the content display interface; The presenting the matching degree corresponding to the second target content in response to the triggering operation on the matching control includes: In response to a matching operation triggered by dragging the second target content from the filter control to the matching control, a matching degree corresponding to the second target content is presented.

16. A content screening method, characterized in that: The method comprises: After receiving an interaction request for a first target content, performing feature analysis on a target object in the first target content to determine at least one key feature information corresponding to the target object, wherein the key feature information is used to filter content matching the target object; Feeding back the at least one key feature information to the client, so that the client presents the at least one key feature information on a content display interface in response to an interactive operation triggered by the first target content, and updates a screening feature set corresponding to the target object based on the at least one key feature information; After receiving a matching request for the second target content, the second target content is compared with at least one key feature information in the screening feature set to obtain a matching result; The matching result is sent to the client, so that the client presents the matching result after responding to the matching operation triggered for the second target content.

17. The method according to claim 16, wherein The performing feature analysis on the target object in the first target content to determine at least one key feature information corresponding to the target object includes: identifying at least one target object contained in the first target content; For each target object, perform the following operations: Collecting hotly discussed information related to a target object from the Internet; and extracting first description information of the target object from the text information corresponding to the first target content; At least one key feature information corresponding to the target object is determined based on the collected hot discussion information and the extracted first description information.

18. The method according to claim 17, wherein The method of collecting hot discussion information related to a target object from the network includes: collecting network data related to the one target object; For each network data, perform the following operations: Extracting a summary of text information corresponding to a piece of network data to generate a text summary corresponding to the piece of network data; Segmenting the text summary into words; For each word, determining the frequency of occurrence of the word in the text summary and the inverse document frequency of the word in network data related to the one target object; Based on the occurrence frequency and inverse document frequency corresponding to each word, hotly discussed information related to the target object is extracted from the text summary.

19. The method according to claim 18, wherein Before extracting a summary of the text information corresponding to a piece of network data to generate a text summary corresponding to the piece of network data, the method further includes: Performing sentiment analysis on the one network data to determine sentiment information related to the one target object in the one network data; extracting comment information related to the target object from the network data; The extracting a summary of text information corresponding to a piece of network data to generate a text summary corresponding to the piece of network data includes: The sentiment information and the comment information are combined to extract a summary of the text information corresponding to a piece of network data, and generate a text summary corresponding to the piece of network data.

20. The method of claim 16, wherein: The matching result includes: a matching degree obtained by comparing the second target content with at least one key feature information in the screening feature set; Then, the second target content is compared with at least one key feature information in the screening feature set to obtain a matching result, including: extracting second description information of the target object from the text information corresponding to the second target content; For each piece of second description information, matching the second description information with at least one key feature information in the screening feature set; The matching degree is determined according to the number of the successfully matched second description information and the total number of the at least one key feature information.

21. A content screening device, characterized in that: include: A presentation unit, configured to present a content display interface; The content display interface is used to display at least one content to be viewed; a first response unit configured to, in response to an interactive operation triggered by a first target content in the at least one content to be viewed, present at least one key feature information corresponding to a target object in the first target content on the content display interface, and update a screening feature set corresponding to the target object based on the at least one key feature information; the key feature information being used to screen content matching the target object; The second response unit is used to present a corresponding matching result in response to a matching operation triggered for a second target content in the at least one content to be viewed; the matching result is: a result obtained by comparing the second target content with at least one key feature information in the screening feature set.

22. A content screening device, characterized in that: include: a first receiving unit configured to, upon receiving an interaction request for a first target content, perform feature analysis on a target object in the first target content to determine at least one key feature information corresponding to the target object, wherein the key feature information is used to filter content matching the target object; a first feedback unit, configured to feed back the at least one key feature information to the client, so that the client presents the at least one key feature information on a content display interface in response to an interactive operation triggered by the first target content, and updates a screening feature set corresponding to the target object based on the at least one key feature information; a second receiving unit, configured to, upon receiving a matching request for a second target content, compare the second target content with at least one key feature information in the screening feature set to obtain a matching result; The second feedback unit is configured to send the matching result to the client, so that the client presents the matching result after responding to the matching operation triggered for the second target content.

23. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor is enabled to perform the steps of the method according to any one of claims 1 to 20.

24. A computer-readable storage medium, characterized in that The method comprises a computer program. When the computer program is run on an electronic device, the computer program is used to enable the electronic device to execute the steps of any one of the methods according to claims 1 to 20.

25. A computer program product, characterized in that The method comprises a computer program stored in a computer-readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the electronic device performs the steps of any one of the methods described in claims 1 to 20.

Images