Abstract generation method and device, equipment and storage medium

Abstract

The invention belongs to the field of computers, and discloses an abstract generation method and device, equipment and a storage medium. The method comprises the steps of obtaining multiple pieces of target information; performing feature extraction on the multiple pieces of target information to obtain information features of the target information; clustering the multiple pieces of target information according to the information features to obtain different types of clustering information; and generating summary information corresponding to the clustering information of each category. As the information to be viewed is clustered, and then the summary information corresponding to the clustered information of each category is generated for the user to view, compared with an existing mode of selecting the target information from the information to be viewed one by one for reading when the user views the information, the target information can be clustered, and the user experience is improved. And the summary information corresponding to the clustering information of each category is generated, so that the reading efficiency of the user is improved.

Description

Technical Field

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

[0002] With the accelerated development of artificial intelligence tools, the efficiency of information processing and production has been improved, and the speed of information output will be much faster than before, which will lead to a significant increase in the amount of information obtained. However, the information obtained by users in various fragmented time may not be timely viewed and understood. When users check information, a large amount of information to be checked may have accumulated. At this time, if users read one by one, it will waste a lot of time. Usually, users will ignore a large amount of unchecked information. Therefore, how to improve users' reading efficiency has become an urgent problem to be solved.

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

[0004] The main purpose of the present invention is to provide a summary generation method, device, equipment and storage medium, aiming to solve the technical problem that the information output speed is fast and the information acquisition amount is greatly improved in the prior art, resulting in low efficiency of user reading information.

[0005] To achieve the above object, the present invention provides a method for generating a summary, the method comprising the following steps:

[0006] Get multiple target information;

[0007] Extracting features of the plurality of target information respectively to obtain information features of each of the target information;

[0008] Clustering the plurality of target information according to the information features to obtain clustering information of different categories;

[0009] Generate summary information corresponding to each category of the clustering information.

[0010] Optionally, after the step of generating summary information corresponding to each category of the clustering information, the step further includes:

[0011] Displaying the clustering information and the corresponding summary information;

[0012] and / or,

[0013] In response to the triggered user query instruction, clustering information and corresponding summary information corresponding to the user query instruction are displayed.

[0014] Optionally, the step of acquiring multiple target information includes:

[0015] In response to the triggered information marking instruction, obtaining the information type corresponding to the information marking instruction;

[0016] When the information type belongs to the preloaded information type, the corresponding target information is acquired according to the information acquisition path in the information marking instruction.

[0017] Optionally, before the step of acquiring multiple target information, the step further includes:

[0018] In response to the triggered information marking instruction, obtaining the information marking condition in the information marking instruction;

[0019] Monitoring the information to be viewed on the target terminal, and determining whether the information to be viewed satisfies the information marking condition;

[0020] When the information to be viewed meets the information marking condition, the information to be viewed is marked as target information.

[0021] Optionally, the step of acquiring multiple target information includes:

[0022] In response to the triggered information viewing instruction, obtaining the information viewing requirement in the information viewing instruction, wherein the information viewing requirement includes the information source, category and / or release time;

[0023] Based on the information viewing requirement, query the information to be viewed and obtain the query result;

[0024] At least one target information to be viewed is acquired according to the query result.

[0025] Optionally, before the step of acquiring multiple target information, the step further includes:

[0026] Monitor user information to view details;

[0027] Determine the information category and viewing status of the target viewing information in the information viewing details;

[0028] When the information category is preset feedback information, determining whether the viewing status is completed;

[0029] When the viewing status is not viewing completion, the target viewing information is used as target information to be viewed.

[0030] Optionally, the step of extracting features from the plurality of target information respectively to obtain information features of each target information includes:

[0031] Performing semantic analysis on the plurality of target information respectively to obtain semantic analysis results;

[0032] Semantic features of the target information are extracted based on the semantic analysis results corresponding to the target information to obtain information features of each target information.

[0033] Optionally, the step of clustering the plurality of target information according to the information features to obtain cluster information of different categories includes:

[0034] Determining the representation entity corresponding to the target information according to the information feature;

[0035] The target information with the same expression entity is clustered to obtain cluster information of different categories.

[0036] Optionally, the step of clustering the plurality of target information according to the information features to obtain cluster information of different categories includes:

[0037] Determining text similarities between pairs of the information features;

[0038] The target information is clustered based on the text similarity to obtain clustering information of different categories.

[0039] Optionally, the step of generating summary information corresponding to each category of the clustering information includes:

[0040] Determine target information corresponding to the clustering information of each category;

[0041] Obtaining target summary information corresponding to the target information;

[0042] The summary information corresponding to the clustering information is generated based on the target summary information.

[0043] Optionally, after the step of generating summary information corresponding to each category of the clustering information, the step further includes:

[0044] Display the summary information, and obtain an information detail viewing instruction triggered by the summary information;

[0045] Determine the target information corresponding to the information details viewing instruction;

[0046] The target information is displayed.

[0047] In addition, to achieve the above-mentioned purpose, the present invention also provides a summary generation device, the device comprising:

[0048] A response module is used to obtain multiple target information;

[0049] A feature extraction module, used to extract features from the plurality of target information respectively to obtain information features of each of the target information;

[0050] A clustering module, used for clustering the plurality of target information according to the information features to obtain clustering information of different categories;

[0051] The summary information generating module is used to generate summary information corresponding to each category of the clustering information.

[0052] In addition, to achieve the above-mentioned purpose, the present invention also proposes a summary generation device, which includes: a memory, a processor, and a summary generation program stored in the memory and executable on the processor, wherein the summary generation program is configured to implement the steps of the summary generation method described above.

[0053] In addition, to achieve the above-mentioned purpose, the present invention further proposes a storage medium, on which a summary generation program is stored, and when the summary generation program is executed by a processor, the steps of the summary generation method described above are implemented.

[0054] The present invention obtains multiple target information; performs feature extraction on the multiple target information respectively to obtain information features of each target information; clusters the multiple target information according to the information features to obtain cluster information of different categories; and generates summary information corresponding to each category of cluster information. Since the present invention clusters the information to be viewed and then generates summary information corresponding to each category of cluster information for the user to view, compared to the existing method in which the user selects target information one by one from the information to be viewed to read when viewing the information, the above method of the present invention can cluster the target information and generate summary information corresponding to each category of cluster information, thereby improving the user's reading efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the structure of a summary generation device in a hardware operating environment involved in an embodiment of the present invention;

[0056] Figure 2 It is a flowchart of the first embodiment of the method for generating a summary of the present invention;

[0057] Figure 3 It is a flowchart of the second embodiment of the method for generating a summary of the present invention;

[0058] Figure 4 It is a flowchart of the third embodiment of the method for generating a summary of the present invention;

[0059] Figure 5 It is a structural block diagram of the first embodiment of the abstract generation device of the present invention.

[0060] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0061] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0062] Reference Figure 1 , Figure 1 The schematic diagram of the structure of the summary generation device of the hardware operating environment involved in the embodiment of the present invention is shown.

[0063] like Figure 1 As shown, the summary generation device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0064] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the summary generating device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0065] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a summary generation program.

[0066] exist Figure 1 In the summary generation device shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the summary generation device of the present invention can be set in the summary generation device, and the summary generation device calls the summary generation program stored in the memory 1005 through the processor 1001, and executes the summary generation method provided by the embodiment of the present invention.

[0067] Based on the above summary generation device, an embodiment of the present invention provides a summary generation method, referring to Figure 2 , Figure 2 FIG. 1 is a flow chart of a first embodiment of a method for generating a summary according to the present invention.

[0068] In this embodiment, the summary generation method includes the following steps:

[0069] Step S10: Acquire multiple target information.

[0070] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a mobile phone, a tablet computer, a personal computer, etc., or an electronic device or information reading device that can realize the above functions. The following takes the information reading device as an example to illustrate this embodiment and the following embodiments.

[0071] It should be noted that the target information may be information to be viewed specified by the user or information to be viewed by the user that is automatically acquired from the network and / or locally based on a preset acquisition strategy. The preset acquisition strategy may be a preset method of acquiring information to be viewed, for example, all information received by the user is used as target information, or information received by the user but not viewed within a preset period of time is used as target information.

[0072] Furthermore, in order to generate summary information of multiple target information when the user views the information and improve the user experience, the step S10 may include:

[0073] Step S100: In response to a triggered information viewing instruction, a plurality of target information to be viewed is obtained.

[0074] It should be noted that the information viewing instruction may be an instruction triggered by a user to view target information to be viewed, and the information viewing instruction may also specify information to be viewed. When there is specified viewing information in the information viewing instruction, the viewing information specified by the user to be viewed is queried from the multiple target information to be viewed and displayed. The multiple target information to be viewed may be information that is detected as not being viewed by the user in a timely manner.

[0075] Furthermore, in order to prevent users from being unable to timely access and understand the information obtained in various fragmented times, and from finding it difficult to find the information they want to read from a large amount of information when they want to view the information, before step S10, the method further includes: responding to an information marking instruction triggered by the user based on the current browsing interface, and recording target information corresponding to the information marking instruction, wherein the target information is the information marked by the user to be viewed.

[0076] It should be noted that when a user receives information to be viewed, and identifies the information as information that needs to be viewed, but does not have enough time to view it or wants to view it later, the user can first mark the information and use it as target information to be viewed. The information marking instruction can be a command triggered by the user to mark the information to be viewed based on the information to be viewed displayed in the current browsing interface, and the information marking instruction is used to use the information to be viewed corresponding to the information marking instruction as the target information to be viewed.

[0077] Furthermore, the information received by the user may be a link, a website, an extraction code or source information of a certain information, and the user needs to extract the information according to the source information of the information. At this time, in order to improve the reading efficiency of the user, the step S10 includes: in response to the triggered information marking instruction, obtaining the information type corresponding to the information marking instruction;

[0078] When the information type belongs to the preloaded information type, the corresponding target information is acquired according to the information acquisition path in the information marking instruction.

[0079] It should be noted that the information type may include preloaded information and direct information. When the information type is direct information, the direct information can be directly used as the target information. The preloaded information may be a web page link, an information extraction code, or the like, which is used to obtain the target information. At this time, when the information type belongs to the preloaded information type, the corresponding target information is obtained according to the information acquisition path in the information marking instruction. For example, the information corresponding to the information marking instruction is: "http: / / 103.64.257.1" because "http: / / 103.64.257.1" is a link to the information. Therefore, in order to avoid the user manually obtaining the information according to the link and then reading it, the information reading device requests the web page information corresponding to the link when the user marks the information as information to be viewed, and uses the web page information as the target information to be viewed.

[0080] Furthermore, if each piece of information received by the user needs to be actively marked before it is added to the target information to be viewed, it will cause cumbersome user operations and low reading efficiency. Therefore, before step S10, it also includes: in response to the triggered information marking instruction, obtaining the information marking condition in the information marking instruction;

[0081] Monitoring the information to be viewed on the target terminal, and determining whether the information to be viewed satisfies the information marking condition;

[0082] When the information to be viewed meets the information marking condition, the information to be viewed is marked as target information.

[0083] It should be noted that the information marking condition may be an information marking rule formulated by a user for certain information that conforms to a certain rule, for example, marking information that has not been read for a long time as information to be viewed, automatically marking information sent by certain senders as target information to be viewed, or extracting text features from the information to be viewed on the target terminal, and automatically marking the information as target information to be viewed when the text features include the text features pre-set in the information marking condition. The target terminal may be an information reading device or a terminal for a user to view information.

[0084] Step S20: extracting features from the plurality of target information respectively to obtain information features of each of the target information.

[0085] It should be noted that the feature extraction of the plurality of target information can be based on a feature extraction algorithm to extract entity features in the target information and / or key features of the text in the target information. The feature extraction algorithm can be a feature extraction algorithm or model capable of extracting text features, such as TF-IDF text feature extraction, a feature extraction model based on word vectors (word2vector), an N-gram model, a bag of words model (Bag of Words, BoW), etc.

[0086] In a specific implementation, the information reading device can deeply understand the contents of the plurality of target information through a large language model, and then the large language model extracts entity features and / or text features in the target information through understanding the target information, thereby obtaining information features of the target information. The large language model can be a pre-trained neural network model that can extract entities or key features in the content based on the understanding of the content.

[0087] Furthermore, in order to ensure that the extracted information features are semantically relevant to the target information, the step S20 may include: performing semantic analysis on the plurality of target information respectively to obtain semantic analysis results;

[0088] Semantic features of the target information are extracted based on the semantic analysis results corresponding to the target information to obtain information features of each target information.

[0089] It should be noted that the semantic analysis of the multiple target information can be performed by performing word segmentation, part-of-speech tagging, dependency analysis, named entity recognition, sentiment analysis, entity relationship recognition, text summarization and reasoning, semantic representation, etc. on the target information to obtain semantic information corresponding to the target information, that is, the semantic analysis result. Among them, the dependency analysis can be to identify the relationship and dependency between each word or phrase. The named entity recognition can be to identify entities such as people, places, organizations, etc. in the text of the target information. The text summarization and reasoning can be to summarize, infer, analyze, etc. based on the content of the text.

[0090] It should be understood that semantic analysis can understand the meaning and context of the text, so as to more accurately extract useful features of the text. Therefore, the semantic feature extraction of the target information based on the semantic analysis result corresponding to the target information can be performed according to the semantics of the target information in the semantic analysis result. Feature extraction, for example, can be performed according to the semantics of the target information in the semantic analysis result. The emotional polarity in the target information can then be extracted as a feature to more accurately classify the target information. It is also possible to first identify named entities in the text through semantic analysis, and extract entity features in the target information based on the named entities, thereby distinguishing different types of text.

[0091] Step S30: clustering the plurality of target information according to the information features to obtain clustering information of different categories.

[0092] It should be noted that clustering the plurality of target information according to the information features may be clustering target information with similar information features into one category according to the information features, thereby obtaining a plurality of clustering information of different categories.

[0093] Further, in order to improve the efficiency of information clustering, the step S30 may include: determining the expression entity corresponding to the target information according to the information feature;

[0094] The target information with the same expression entity is clustered to obtain cluster information of different categories.

[0095] It should be noted that the expression entity may be an entity with a specific meaning in the target information, such as a name, a place name, an organization name, a time, etc. The entity with a specific meaning in the target information may be determined based on the information feature. Then, the target information with the same entity is clustered into one category to obtain clustering information of different categories. Among them, the target information may be clustered by the following clustering algorithms: K-means, hierarchical clustering, DBSCAN, etc.

[0096] In a specific implementation, the information reading device inputs the information features into a large language model. The large language model can determine the expression entities corresponding to the target information based on the information features, and cluster the target information according to the correlation between the expression entities corresponding to the target information to obtain clustering information of different categories.

[0097] Furthermore, the target information may also be clustered directly according to the text similarity of the information features, and the step S30 may include: determining the text similarity between each pair of the information features;

[0098] The target information is clustered based on the text similarity to obtain clustering information of different categories.

[0099] It should be noted that the text similarity may be the semantic similarity of the information features of the target information, or may be cosine similarity, Jaccard similarity, etc. Clustering the target information based on the text similarity may be clustering the target information corresponding to the information features whose text similarity is greater than a preset similarity threshold. It may also be clustering the target information using a clustering algorithm such as K-Means, hierarchical clustering, DBSCAN, etc. to obtain clustering information of different categories.

[0100] Step S40: generating summary information corresponding to each category of the clustering information.

[0101] It should be noted that the generation of summary information corresponding to the cluster information of each category may be to input the cluster information of each category into the large language model, and then use the large language model to compress the content of the cluster information in the same category and summarize it into concise summary information. For example, summary information of the same cluster information may also be generated in the following ways: extractive summarization, generative summarization, and hybrid summarization.

[0102] Furthermore, in order to facilitate the user to view the summary information and clustering information corresponding to the target information, after step S40, the following is further included:

[0103] Displaying the clustering information and the corresponding summary information;

[0104] and / or,

[0105] In response to the triggered user query instruction, clustering information and corresponding summary information corresponding to the user query instruction are displayed.

[0106] It should be noted that the user query instruction may be a command triggered by the user to view a certain type of clustering information and / or summary information corresponding to a certain type of clustering information. The display of the clustering information and the corresponding summary information may be displaying the clustering information and the corresponding summary information according to a preset display strategy. For example, the preset display strategy may be to display the summary information corresponding to each category of clustering information from top to bottom in the interface of the information reading device, and after the summary information is displayed, the clustering information of each category is displayed. It is also possible to display the clustering information and the corresponding summary information of a certain category first, and then display the clustering information and the corresponding summary information of other categories, etc. This embodiment is not limited here.

[0107] Furthermore, in order to more accurately generate summary information corresponding to each category of cluster information, the step S40 may include: determining target information corresponding to each category of cluster information;

[0108] Obtaining target summary information corresponding to the target information;

[0109] The summary information corresponding to the clustering information is generated based on the target summary information.

[0110] It should be noted that the target summary information corresponding to the target information may be the original summary information in the target information. If the target information is subscription information, the target summary information may be the summary information in the subscription information; if the target information is email information, the target summary information may be the body information in the email information; if the target information is a notification message, the target summary information may be the subject information in the notification message. The generation of the summary information corresponding to the clustering information based on the target summary information may be based on the understanding of the target summary information by the large language model, and the summary information corresponding to the clustering information may also be generated based on the length of the type of summary set by the user. For example, if the type of summary information output selected by the user is an audio and video type, the summary audio corresponding to the clustering information is output as the summary information. If the type of summary information output selected by the user is a text type, the length of the text type summary information set by the user is obtained, and the summary information not longer than the length of the text type summary information is output.

[0111] Furthermore, in order to facilitate the user to view the original complete information corresponding to the summary information, after the step S40, it may include: displaying the summary information, and obtaining an information detail viewing instruction triggered by the summary information;

[0112] Determine the target information corresponding to the information details viewing instruction;

[0113] The target information is displayed.

[0114] It should be noted that the information details viewing instruction may be an instruction for viewing original information corresponding to the summary information triggered based on the displayed summary information.

[0115] In a specific implementation, the summary information displayed to the user may be obtained by simplifying and extracting multiple target information from different sources. Therefore, when the user triggers an information detail viewing instruction to view the original information corresponding to the summary information, it is necessary to query which target information the target summary comes from according to the target summary specified in the information detail viewing instruction, and display the target information.

[0116] This embodiment obtains multiple target information; extracts features from the multiple target information respectively to obtain information features of each target information; clusters the multiple target information according to the information features to obtain cluster information of different categories; generates summary information corresponding to each category of cluster information. Since this embodiment clusters the information to be viewed, and then generates summary information corresponding to each category of cluster information for the user to view, compared with the existing method in which the user selects target information one by one from the information to be viewed when viewing the information, the above method of this embodiment can cluster the target information and generate summary information corresponding to each category of cluster information, thereby improving the user's reading efficiency.

[0117] refer to Figure 3 , Figure 3 FIG. 4 is a flow chart of a second embodiment of a method for generating a summary according to the present invention.

[0118] Based on the first embodiment above, in this embodiment, before step S10, the following steps are further included:

[0119] Step S001: Monitor user information and view details.

[0120] It should be noted that the information viewing details may include the information category and viewing status of the information currently being viewed by the user, and the viewing status may include whether the user has completed viewing the current information. For example, when a user views information through an information reading device, the user's viewing behavior is monitored, and it is determined based on the user's viewing behavior whether the user has browsed to the end of the currently viewed information. If the user has browsed to the end of the currently viewed information, it is determined that the user has completed viewing the information.

[0121] Step S002: Determine the information category and viewing status of the target viewing information in the information viewing details.

[0122] It should be noted that the information category may include reading information and preset feedback information. The target viewing information may be information included in the information viewing details. The preset feedback information may be information that requires a user to reply after reading, such as email information, personal chat information of a communication program, etc. The reading information may include subscription information, push information, and advertising information, etc.

[0123] Step S003: When the information category is preset feedback information, determine whether the viewing status is completed.

[0124] In a specific implementation, when the information category is preset feedback information, the information reading device determines whether the viewing status is completed, and does not process when the information category is reading information.

[0125] Step S004: When the viewing status is not viewing completed, the target viewing information is used as the target information to be viewed.

[0126] It should be understood that when the viewing status is not completed, it means that the user has not completed viewing the preset feedback information, which may cause the user to miss the preset feedback information. Therefore, in this embodiment, when the viewing status is not completed, the target viewing information is used as the target information to be viewed to remind the user to view it again. When the viewing status is completed, in order to prevent the user from forgetting to reply, a reminder time can also be set. When the reminder time is reached, the user is reminded to reply to the preset feedback information.

[0127] This embodiment monitors the user's information viewing details; determines the information category and viewing status of the target viewing information in the information viewing details; when the information category is preset feedback information, determines whether the viewing status is completed viewing; when the viewing status is not completed viewing, the target viewing information is used as the target information to be viewed. This embodiment monitors the user's information viewing details and determines whether the user has completed viewing the preset feedback information, thereby assisting the user in viewing the preset feedback information, preventing the user from missing the processing of the preset feedback information, and improving the user's reading experience.

[0128] refer to Figure 4 , Figure 4 FIG. 4 is a flow chart of a third embodiment of a method for generating a summary according to the present invention.

[0129] Based on the above embodiments, in this embodiment, step S10 includes:

[0130] Step S101: in response to a triggered information viewing instruction, obtaining information viewing requirements in the information viewing instruction, wherein the information viewing requirements include information source, category and / or release time.

[0131] It should be noted that the information viewing requirement may include relevant information of the information that the user wants to view, such as the information subject, name, source, category, information format and / or release time.

[0132] Step S102: query the information to be viewed based on the information viewing requirement to obtain the query result.

[0133] It should be noted that the query result obtained by querying the information to be viewed based on the information viewing requirement can be to query the information to be viewed that meets the information viewing requirement according to the information viewing requirement and generate the query result. For example, if the information viewing requirement is email information, the email information in the information to be viewed is queried and the query result is generated based on the queried email information.

[0134] Step S103: obtaining at least one target information to be viewed according to the query result.

[0135] It should be noted that, acquiring at least one target information to be viewed according to the query result may be taking the information in the query result as the target information to be viewed.

[0136] In response to the triggered information viewing instruction, this embodiment obtains the information viewing requirement in the information viewing instruction, and the information viewing requirement includes the information source, category and / or release time; queries the information to be viewed based on the information viewing requirement to obtain the query result; and obtains at least one target information to be viewed according to the query result. This embodiment can filter the information to be viewed according to the information viewing requirement in the user information viewing instruction, obtain the target information that meets the user's viewing requirement, and improve the user's reading experience.

[0137] Reference Figure 5 , Figure 5 It is a structural block diagram of the first embodiment of the abstract generation device of the present invention.

[0138] like Figure 5 As shown, the summary generation device proposed in the embodiment of the present invention includes:

[0139] A response module 10 is used to obtain multiple target information;

[0140] A feature extraction module 20 is used to extract features from the plurality of target information to obtain information features of each target information;

[0141] A clustering module 30, configured to cluster the plurality of target information according to the information features to obtain clustering information of different categories;

[0142] The summary information generating module 40 is used to generate summary information corresponding to each category of the clustering information.

[0143] This embodiment obtains multiple target information; extracts features from the multiple target information respectively to obtain information features of each target information; clusters the multiple target information according to the information features to obtain cluster information of different categories; generates summary information corresponding to each category of cluster information. Since this embodiment clusters the information to be viewed, and then generates summary information corresponding to each category of cluster information for the user to view, compared with the existing method in which the user selects target information one by one from the information to be viewed when viewing the information, the above method of this embodiment can cluster the target information and generate summary information corresponding to each category of cluster information, thereby improving the user's reading efficiency.

[0144] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0145] In addition, for technical details not fully described in this embodiment, reference may be made to the summary generation method provided in any embodiment of the present invention, and will not be repeated here.

[0146] Based on the first embodiment of the abstract generation device of the present invention, a second embodiment of the abstract generation device of the present invention is proposed.

[0147] In this embodiment, the summary information generating module 40 is also used to display the clustering information and the corresponding summary information;

[0148] and / or,

[0149] In response to the triggered user query instruction, clustering information and corresponding summary information corresponding to the user query instruction are displayed.

[0150] Furthermore, the response module 10 is further configured to obtain the information type corresponding to the information marking instruction in response to the triggered information marking instruction;

[0151] When the information type belongs to the preloaded information type, the corresponding target information is acquired according to the information acquisition path in the information marking instruction.

[0152] Furthermore, the response module 10 is further configured to obtain the information marking condition in the information marking instruction in response to the triggered information marking instruction;

[0153] Monitoring the information to be viewed on the target terminal, and determining whether the information to be viewed satisfies the information marking condition;

[0154] When the information to be viewed meets the information marking condition, the information to be viewed is marked as target information.

[0155] Furthermore, the response module 10 is further configured to obtain, in response to the triggered information viewing instruction, an information viewing requirement in the information viewing instruction, wherein the information viewing requirement includes information source, category and / or release time;

[0156] Based on the information viewing requirement, query the information to be viewed and obtain the query result;

[0157] At least one target information to be viewed is acquired according to the query result.

[0158] Furthermore, the response module is also used to monitor the user's information viewing details;

[0159] Determine the information category and viewing status of the target viewing information in the information viewing details;

[0160] When the information category is preset feedback information, determining whether the viewing status is completed;

[0161] When the viewing status is viewing completed, the target viewing information is used as the target information to be viewed.

[0162] Furthermore, the response module is further used to perform semantic analysis on the plurality of target information respectively to obtain semantic analysis results;

[0163] Semantic features of the target information are extracted based on the semantic analysis results corresponding to the target information to obtain information features of each target information.

[0164] Furthermore, the clustering module 30 is further used to determine the expression entity corresponding to the target information according to the information feature;

[0165] The target information with the same expression entity is clustered to obtain cluster information of different categories.

[0166] Furthermore, the clustering module 30 is also used to determine the text similarity between the information features;

[0167] The target information is clustered based on the text similarity to obtain clustering information of different categories.

[0168] Furthermore, the summary information generating module 40 is also used to determine the target information corresponding to the clustering information of each category;

[0169] Obtaining target summary information corresponding to the target information;

[0170] The summary information corresponding to the clustering information is generated based on the target summary information.

[0171] Furthermore, the summary information generating module 40 is further used to display the summary information and obtain an information detail viewing instruction triggered by the summary information;

[0172] Determine the target information corresponding to the information details viewing instruction;

[0173] The target information is displayed.

[0174] Other embodiments or specific implementations of the summary generation device of the present invention may refer to the above-mentioned method embodiments, which will not be described in detail here.

[0175] In addition, an embodiment of the present invention further provides a storage medium, on which a summary generation program is stored. When the summary generation program is executed by a processor, the steps of the summary generation method described above are implemented.

[0176] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0177] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0178] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0179] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

[0180] The present invention discloses A1, a method for generating a summary, the method for generating a summary comprising the following steps:

[0181] Get multiple target information;

[0182] Extracting features of the plurality of target information respectively to obtain information features of each of the target information;

[0183] Clustering the plurality of target information according to the information features to obtain clustering information of different categories;

[0184] Generate summary information corresponding to each category of the clustering information.

[0185] A2. The summary generation method as described in A1, after the step of generating summary information corresponding to the clustering information of each category, further comprising:

[0186] Displaying the clustering information and the corresponding summary information;

[0187] and / or,

[0188] In response to the triggered user query instruction, clustering information and corresponding summary information corresponding to the user query instruction are displayed.

[0189] A3. In the summary generation method as described in A1, the step of obtaining multiple target information comprises:

[0190] In response to the triggered information marking instruction, obtaining the information type corresponding to the information marking instruction;

[0191] When the information type belongs to the preloaded information type, the corresponding target information is acquired according to the information acquisition path in the information marking instruction.

[0192] A4. The method for generating a summary as described in A1, before the step of obtaining a plurality of target information, further comprising:

[0193] In response to the triggered information marking instruction, obtaining the information marking condition in the information marking instruction;

[0194] Monitoring the information to be viewed on the target terminal, and determining whether the information to be viewed satisfies the information marking condition;

[0195] When the information to be viewed meets the information marking condition, the information to be viewed is marked as target information.

[0196] A5. In the summary generation method as described in A1, the step of obtaining multiple target information comprises:

[0197] In response to the triggered information viewing instruction, obtaining the information viewing requirement in the information viewing instruction, wherein the information viewing requirement includes the information source, category and / or release time;

[0198] Based on the information viewing requirement, query the information to be viewed and obtain the query result;

[0199] At least one target information to be viewed is acquired according to the query result.

[0200] A6. The method for generating a summary as described in A1, before the step of obtaining a plurality of target information, further comprising:

[0201] Monitor user information to view details;

[0202] Determine the information category and viewing status of the target viewing information in the information viewing details;

[0203] When the information category is preset feedback information, determining whether the viewing status is completed;

[0204] When the viewing status is not viewing completion, the target viewing information is used as target information to be viewed.

[0205] A7. In the summary generation method as described in any one of A1 to A6, the step of extracting features from the plurality of target information to obtain information features of each target information comprises:

[0206] Performing semantic analysis on the plurality of target information respectively to obtain semantic analysis results;

[0207] Semantic features of the target information are extracted based on the semantic analysis results corresponding to the target information to obtain information features of each target information.

[0208] A8. In the summary generation method as described in any one of A1 to A6, the step of clustering the plurality of target information according to the information features to obtain cluster information of different categories comprises:

[0209] Determining the representation entity corresponding to the target information according to the information feature;

[0210] The target information with the same expression entity is clustered to obtain cluster information of different categories.

[0211] A9. In the summary generation method as described in any one of A1 to A6, the step of clustering the plurality of target information according to the information features to obtain cluster information of different categories comprises:

[0212] Determining text similarities between pairs of the information features;

[0213] The target information is clustered based on the text similarity to obtain clustering information of different categories.

[0214] A10. In the summary generation method as described in any one of A1 to A6, the step of generating summary information corresponding to each category of the clustering information comprises:

[0215] Determine target information corresponding to the clustering information of each category;

[0216] Obtaining target summary information corresponding to the target information;

[0217] The summary information corresponding to the clustering information is generated based on the target summary information.

[0218] A11. The method for generating a summary as described in any one of A1 to A7, after the step of generating summary information corresponding to the clustering information of each category, further comprising:

[0219] Display the summary information, and obtain an information detail viewing instruction triggered by the summary information;

[0220] Determine the target information corresponding to the information details viewing instruction;

[0221] The target information is displayed.

[0222] The present invention discloses B12, a summary generation device, the summary generation device comprising:

[0223] A response module is used to obtain multiple target information;

[0224] A feature extraction module, used to extract features from the plurality of target information respectively to obtain information features of each of the target information;

[0225] A clustering module, used for clustering the plurality of target information according to the information features to obtain clustering information of different categories;

[0226] The summary information generating module is used to generate summary information corresponding to each category of the clustering information.

[0227] B13. The summary generating device as described in B12, wherein the summary information generating module is further used to display the clustering information and the corresponding summary information;

[0228] and / or,

[0229] In response to the triggered user query instruction, clustering information and summary information corresponding to the user query instruction are displayed.

[0230] B14. The summary generating device as described in B14, wherein the response module is further used to respond to the triggered information marking instruction and obtain the information type corresponding to the information marking instruction;

[0231] When the information type belongs to the preloaded information type, the corresponding target information is acquired according to the information acquisition path in the information marking instruction.

[0232] B15. The summary generating device as described in B12, wherein the response module is further used to obtain the information marking condition in the information marking instruction in response to the triggered information marking instruction;

[0233] Monitoring the information to be viewed on the target terminal, and determining whether the information to be viewed satisfies the information marking condition;

[0234] When the information to be viewed meets the information marking condition, the information to be viewed is marked as target information.

[0235] B16. The summary generating device as described in B12, wherein the response module is further used to obtain the information viewing requirement in the information viewing instruction in response to the triggered information viewing instruction, wherein the information viewing requirement includes the information source, category and / or release time;

[0236] Based on the information viewing requirement, query the information to be viewed and obtain the query result;

[0237] At least one target information to be viewed is acquired according to the query result.

[0238] B17, the summary generation device as described in B12, wherein the response module is further used to monitor the user's information viewing details;

[0239] Determine the information category and viewing status of the target viewing information in the information viewing details;

[0240] When the information category is preset feedback information, determining whether the viewing status is completed;

[0241] When the viewing status is not viewing completion, the target viewing information is used as target information to be viewed.

[0242] B18. The abstract generation device according to any one of B12 to B17, wherein the feature extraction module is further configured to perform semantic analysis on the plurality of target information to obtain semantic analysis results;

[0243] Semantic features of the target information are extracted based on the semantic analysis results corresponding to the target information to obtain information features of each target information.

[0244] The present invention discloses C19, a summary generation device, comprising: a memory, a processor, and a summary generation program stored in the memory and executable on the processor, wherein the summary generation program is configured to implement the steps of the summary generation method described in any one of A1 to A11.

[0245] The present invention discloses D20, a storage medium, on which a summary generation program is stored. When the summary generation program is executed by a processor, the steps of the summary generation method as described in any one of A1 to A11 are implemented.

Claims

1. A method for generating a summary, characterized in that: The summary generation method comprises the following steps: Get multiple target information; Extracting features of the plurality of target information respectively to obtain information features of each of the target information; Clustering the plurality of target information according to the information features to obtain clustering information of different categories; Generate summary information corresponding to each category of the clustering information.

2. The method for generating a summary according to claim 1, wherein: After the step of generating summary information corresponding to the clustering information of each category, the method further includes: Displaying the clustering information and the corresponding summary information; and / or, In response to the triggered user query instruction, clustering information and corresponding summary information corresponding to the user query instruction are displayed.

3. The method for generating a summary according to claim 1, wherein: The step of obtaining multiple target information includes: In response to the triggered information marking instruction, obtaining the information type corresponding to the information marking instruction; When the information type belongs to the preloaded information type, the corresponding target information is acquired according to the information acquisition path in the information marking instruction.

4. The method for generating a summary according to claim 1, wherein: Before the step of obtaining multiple target information, the method further includes: In response to the triggered information marking instruction, obtaining the information marking condition in the information marking instruction; Monitoring the information to be viewed on the target terminal, and determining whether the information to be viewed satisfies the information marking condition; When the information to be viewed meets the information marking condition, the information to be viewed is marked as target information.

5. The method for generating a summary according to claim 1, wherein: The step of obtaining multiple target information includes: In response to the triggered information viewing instruction, obtaining the information viewing requirement in the information viewing instruction, wherein the information viewing requirement includes the information source, category and / or release time; Based on the information viewing requirement, query the information to be viewed and obtain the query result; At least one target information to be viewed is acquired according to the query result.

6. The method for generating a summary according to claim 1, wherein: Before the step of obtaining multiple target information, the method further includes: Monitor user information to view details; Determine the information category and viewing status of the target viewing information in the information viewing details; When the information category is preset feedback information, determining whether the viewing status is completed; When the viewing status is not viewing completion, the target viewing information is used as target information to be viewed.

7. The method for generating a summary according to any one of claims 1 to 6, wherein: The step of extracting features from the plurality of target information respectively to obtain information features of each target information comprises: Performing semantic analysis on the plurality of target information respectively to obtain semantic analysis results; Semantic features of the target information are extracted based on the semantic analysis results corresponding to the target information to obtain information features of each target information.

8. A summary generation device, characterized in that: The summary generating device comprises: A response module is used to obtain multiple target information; A feature extraction module, used to extract features from the plurality of target information respectively to obtain information features of each of the target information; A clustering module, used for clustering the plurality of target information according to the information features to obtain clustering information of different categories; The summary information generating module is used to generate summary information corresponding to each category of the clustering information.

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

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

Images