Event context generation method and device, equipment and medium
By extracting and generating event triplets, the problem of users' difficulty in tracing the development of hot events is solved, the generation of event context is achieved, and the user experience is improved.
Patent Information
- Application Number
- CN202310781208.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-28
- Publication Date
- 2024-12-31
AI Technical Summary
It is difficult for users to effectively trace and understand the development process of hot events on the Internet, and there is a lack of effective method of generating events contexts.
Provide an event context generation method. By obtaining the target event, if it is a hot event, the event triple (event occurrence time, event subject, event information) is extracted from its corresponding information group, and the event context is generated based on this information. .
The event context generation of hot events is realized, allowing users to understand the development process of events by browsing event contexts, and improving users' experience of watching information.
Smart Images
Figure CN119226520A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of vehicle technology, and in particular to an event context generation method, device, equipment and medium. Background Art
[0002] In recent years, the Internet has become an important means for users to acquire knowledge and information, and has also become a platform for the collection and fermentation of diverse information.
[0003] Usually, when users obtain information on the Internet and are interested in it, they often want to know the entire process of the development of the event in the information to fully understand the ins and outs of the event. Therefore, there is an urgent need for a method to trace the context of events to meet user needs. Summary of the invention
[0004] In order to solve the above technical problems, the present disclosure provides an event context generation method, device, equipment and medium.
[0005] In a first aspect, the present disclosure provides an event context generation method, comprising:
[0006] Get the target event;
[0007] If the target event is a hot event, then extract the event triple of the target event from the information group corresponding to the target event, wherein the event triple includes the event occurrence time, event subject and event information;
[0008] The target event is tracked according to its event triplet to generate event context.
[0009] In a second aspect, the present disclosure provides an event context generation device, comprising:
[0010] An acquisition module is used to acquire target events;
[0011] An extraction module, for extracting an event triple of the target event from an information group corresponding to the target event if the target event is a hot event, wherein the event triple includes the event occurrence time, the event subject, and the event information;
[0012] The generation module is used to track the target event according to the event triple of the target event and generate the event context.
[0013] In a third aspect, the present disclosure provides an electronic device, including:
[0014] processor;
[0015] A memory for storing executable instructions;
[0016] Among them, the processor is used to read executable instructions from the memory and execute the executable instructions to implement the event context generation method described in the first aspect.
[0017] In a fourth aspect, the present disclosure provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor implements the event context generation method described in the first aspect.
[0018] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:
[0019] The event context generation method, device, equipment and medium of the disclosed embodiment can obtain the target event; if the target event is a hot event, the event triple of the target event is extracted from the information group corresponding to the target event, wherein the event triple includes the time of event occurrence, event subject and event information; the target event is tracked according to the event triple of the target event to generate the event context. It can be seen that the event context of the hot event can be generated by adopting the above technical solution, so that the user can understand the development process of the hot event by browsing the event context of the hot event, which is conducive to improving the user's experience of viewing information. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.
[0021] Figure 1 A schematic diagram of a process of generating an event context provided by an embodiment of the present disclosure is shown;
[0022] Figure 2 A schematic diagram showing a flow chart of another event context generation method provided by an embodiment of the present disclosure;
[0023] Figure 3 A schematic diagram of the structure of an event context generating device provided by an embodiment of the present disclosure is shown;
[0024] Figure 4 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0025] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.
[0026] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.
[0027] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.
[0028] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.
[0029] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".
[0030] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.
[0031] In order to solve the above problems, the present disclosure provides an event context generation method, device, equipment and medium. The event context generation method provided by the present disclosure is first described below.
[0032] Figure 1 A schematic diagram of a flow chart of an event context generation method provided by an embodiment of the present disclosure is shown.
[0033] In some embodiments of the present disclosure, Figure 1The event context generation method shown can be applied to an event context generation device. The event context generation device can be implemented by software and / or hardware and integrated into an electronic device. The electronic device described here can include, for example, a computer, a notebook, a mobile phone, a server, etc., without specific limitation.
[0034] like Figure 1 As shown, the event context generation method may include the following steps.
[0035] S110: Obtain target event.
[0036] Specifically, the target event is any event whose event context is to be tracked.
[0037] In some embodiments, S110 may include:
[0038] S1111. Acquire from multiple data sources a plurality of information that is published within a first preset time period and belongs to a target field.
[0039] Specifically, there are many ways to determine the first preset time period, which are described below with reference to typical examples, but are not limited thereto.
[0040] In one example, the first preset time period can be determined by the current time and the first preset duration. For example, the current time can be used as the end time of the first preset time period, and the time obtained by tracing back from the current time to the first preset duration can be used as the start time of the first preset time period, wherein the specific value of the first preset duration can be set by technicians in this field according to actual conditions and is not limited here. For example, if the current time is 8:00 a.m. on May 10, 2023, and the first preset duration is one day, then the first preset time period is from 8:00 a.m. on May 9, 2023 to 8:00 a.m. on May 10, 2023.
[0041] In another example, in response to a first setting operation of a user on a first preset time period, a first preset time period corresponding to the first setting operation may be determined. For example, a user may input a specific value of the first preset time period (i.e., a first setting operation) through a mouse, keyboard, etc., such as 00:00 on June 26, 2023 - 00:00 on June 27, 2023, and the electronic device may respond to the first setting operation and set 00:00 on June 26, 2023 - 00:00 on June 27, 2023 as the first preset time period.
[0042] Specifically, those skilled in the art may set the target field according to the actual field of interest. For example, the target field may include the vehicle field, the computer field, or the game field, but is not limited thereto.
[0043] Specifically, the data source may include the Internet, or a designated location, etc. Accordingly, the way to obtain information may include crawling from the Internet using a crawler, or reading from a designated location (such as a local storage, a cloud, etc.), etc. But it is not limited thereto.
[0044] It is understandable that filtering information from multiple data sources based on release time and target fields can make the acquired information more timely and more in line with the areas of user concern, which is conducive to obtaining effective target events.
[0045] S1112. Cluster multiple pieces of information to obtain at least one information group.
[0046] Specifically, clustering algorithms such as DBSCAN may be used for clustering, but are not limited thereto.
[0047] Specifically, by clustering multiple pieces of information, similar information (ie, information about the same event) can be divided into the same information group, so each information group corresponds to an event.
[0048] S1113: If the amount of information included in the information group is greater than a first preset threshold, determine that the event corresponding to the information group is a target event.
[0049] Specifically, the specific value of the first preset threshold can be set by those skilled in the art according to actual conditions and is not limited here.
[0050] Specifically, each information group corresponds to a cluster center (i.e., information that can better represent the event corresponding to the information group). By performing topic extraction on the cluster center, the event corresponding to the information group can be obtained. Those skilled in the art can use any possible topic extraction algorithm to perform topic extraction on the cluster center without limitation. For example, performing topic extraction on the cluster center of a certain information group can obtain "#XX Car Enterprise L9 Release Conference#" and the like.
[0051] It is understandable that if the amount of information contained in the information group is greater than the first preset threshold, it indicates that the event corresponding to the information group has a certain degree of attention, and taking it as the target event can make the acquired target event an event that the user is more concerned about.
[0052] In some other embodiments, S110 may further include:
[0053] S1121. Select any data source from multiple data sources as a target data source, and obtain from the target data source a plurality of information that is published within a first preset time period and belongs to a target field.
[0054] Specifically, any data source may be selected as the target data source, or the data source with the largest amount of historical information may be selected as the target data source, but the present invention is not limited thereto.
[0055] S1122. Cluster multiple pieces of information to obtain at least one sub-information group originating from a target data source.
[0056] S1123: If the amount of information included in the sub-information group is greater than a fourth preset threshold, determine that the event corresponding to the sub-information group is a target event.
[0057] Specifically, the specific value of the fourth preset threshold can be set by those skilled in the art according to actual conditions and is not limited here.
[0058] It can be understood that acquiring the target event based on the information in the target data source can reduce the amount of information referenced in the process of acquiring the target event, which is beneficial to improving the efficiency of acquiring the target event.
[0059] S120: If the target event is a hot event, extracting an event triplet of the target event from the information group corresponding to the target event, wherein the event triplet includes the event occurrence time, the event subject, and the event information.
[0060] Specifically, hot events are events that attract widespread attention in society, participate in discussions, stir up public emotions, and trigger strong reactions.
[0061] Specifically, there are many specific ways to detect whether a target event is a hot target event, which are described below with reference to typical examples, but are not limited thereto.
[0062] In some embodiments, the information group corresponding to the target event includes at least one sub-information group, and different sub-information groups are derived from different data sources; wherein the detection of whether the target event is a hot target event includes:
[0063] S121. For each sub-information group, count the index parameters of the sub-information group, and analyze the probability value of the target event being a hot event based on the index parameters of the sub-information group to obtain the probability value corresponding to the sub-information group, wherein the index parameters include the number of readings, the number of likes, the number of comments, the number of reposts, and the number of collections.
[0064] Specifically, the index parameters (reading volume, likes, comments, forwarding volume, and collection volume) of a sub-information group refer to the sum of the index parameters (reading volume, likes, comments, forwarding volume, and collection volume) of each information in the sub-information group. For example, the reading volume of a sub-information group is the sum of the reading volume of each information in the sub-information group, and the same is true for the likes, comments, forwarding, and collection volume, which will not be repeated here.
[0065] In one example, the probability value of the target event being a hot event is analyzed based on the indicator parameters of the sub-information group to obtain the probability value corresponding to the sub-information group, including: weighted summation of the reading volume, like volume, comment volume, forwarding volume, and collection volume of the sub-information group to obtain the probability value corresponding to the sub-information group.
[0066] In another example, the probability value of the target event being a hot event is analyzed according to the indicator parameters of the sub-information group to obtain the probability value corresponding to the sub-information group, including:
[0067] The weighted sum of the number of likes, comments, reposts, and favorites of the sub-information group is calculated to obtain the sum value;
[0068] A logarithmic operation is performed based on the reading volume and the sum of the sub-information groups to obtain a probability value corresponding to the sub-information group, wherein the base of the logarithmic operation is determined by the reading volume of the sub-information group and the true number of the logarithmic operation is determined by the sum.
[0069] Specifically, the weight values corresponding to the number of likes, comments, reposts, and favorites of the sub-information group can be set by those skilled in the art according to actual conditions and are not limited here.
[0070] Specifically, the base of the logarithmic operation can be the reading volume of the sub-information group, or the product of the reading volume of the sub-information group and the first preset coefficient, etc., and the true number of the logarithmic operation can be the sum, or the product of the sum and the second preset coefficient, etc., but is not limited to this.
[0071] S122. Add the probability values corresponding to each sub-information group to obtain a total probability value.
[0072] S123: If the total probability value is greater than a second preset threshold, determine that the target event is a hot event.
[0073] Specifically, the specific value of the second preset threshold can be set by those skilled in the art according to actual conditions and is not limited here. For example, the second preset threshold can be 0.5, but is not limited thereto.
[0074] Exemplarily, for any target event, the calculation formula for the total probability value of the target event can be as follows:
[0075]
[0076] Among them, P represents the probability that the target event is a hot event, i represents the sequence number of the i-th data source, there are n data sources in total, R i Indicates the reading volume of the sub-information group corresponding to the target event from the i-th data source, like i Indicates the number of likes in the sub-information group corresponding to the target event from the i-th data source; comment iIndicates the number of comments in the sub-information group corresponding to the target event from the i-th data source; forward i Indicates the forwarding volume of the sub-information group corresponding to the target event from the i-th data source; favorite i Indicates the number of sub-information group collections corresponding to the target event from the ith data source; a, b, c, d are the weights of the corresponding indicator parameters. For example, a = 0.2, b = 0.1, c = 0.5, d = 0.3, but not limited to this.
[0077] Of course, the probability values corresponding to each sub-information group can also be weighted and summed to obtain a total weighted probability value. If the total weighted probability value is greater than the fifth preset threshold, the target event is determined to be a hot event. The specific value of the fifth preset threshold can be set by those skilled in the art according to actual conditions and is not limited here.
[0078] It can be understood that by separately determining the probability value corresponding to each sub-information group and judging whether the target event is a hot event based on the probability value corresponding to each sub-information group, it is possible to separately determine the popularity of the target event in each data source and comprehensively judge whether the target event is a hot event based on the popularity in each data source, which is conducive to improving the accuracy of judging whether the target event is a hot event.
[0079] Specifically, the event occurrence time is the time when the event occurs. The event subject is the object or thing related to the event. The event information is the information reflecting the main content of the event.
[0080] Specifically, there are many specific implementation methods for extracting the event triple of the target event from the information group corresponding to the target event. A typical example is described below, but the invention is not limited thereto.
[0081] In some embodiments, extracting the event triple of the target event from the information group corresponding to the target event may include: determining the cluster center of the information group corresponding to the target event; performing word segmentation processing and part-of-speech analysis processing on the cluster center to obtain multiple participles of the cluster center and the parts-of-speech of the multiple participles; and extracting the event triple of the target event from the multiple participles according to the parts-of-speech of the multiple participles. However, the present invention is not limited thereto.
[0082] In other embodiments, extracting the event triplet of the target event from the information group corresponding to the target event may include: inputting the cluster center of the information group corresponding to the target event (or the information group corresponding to the target event) into the event triplet extraction model to obtain the event triplet of the target event output by the event triplet extraction model, wherein the event triplet extraction model is pre-trained based on the sample information (or sample information group) and the event triplet of the sample information (or sample information group).
[0083] S130: Track the target event according to the event triplet of the target event to generate an event context.
[0084] Specifically, the event context is a context formed by sorting the event triples of multiple events in the order of the occurrence time of the corresponding events.
[0085] Specifically, according to the event triplet of the target event, the target event is tracked to obtain the key event corresponding to the target event, wherein the correlation between the key event and the target event meets the preset conditions and is a hot event; the event triplet of the key event is extracted from the information group corresponding to the key event, and the event context is generated according to the event triplet of the target event and the event triplet of the key event.
[0086] The event context generation method, device, equipment and medium of the disclosed embodiment can obtain the target event; if the target event is a hot event, the event triple of the target event is extracted from the information group corresponding to the target event, wherein the event triple includes the time of event occurrence, event subject and event information; the target event is tracked according to the event triple of the target event to generate the event context. It can be seen that the event context of the hot event can be generated by adopting the above technical solution, so that the user can understand the development process of the hot event by browsing the event context of the hot event, which is conducive to improving the user's experience of viewing information.
[0087] Figure 2 The flowchart of another event context generation method provided by the embodiment of the present disclosure is shown. The embodiment of the present disclosure is optimized on the basis of the above embodiment, and the embodiment of the present disclosure can be combined with various optional solutions in one or more of the above embodiments.
[0088] like Figure 2 As shown, the event context generation method may include the following steps.
[0089] S210: Obtain target event.
[0090] Specifically, S210 is similar to S110 and will not be described in detail here.
[0091] S220: If the target event is a hot event, extract the event triple of the target event from the information group corresponding to the target event, wherein the event triple includes the event occurrence time, the event subject, and the event information.
[0092] Specifically, S220 is similar to S120 and will not be described in detail here.
[0093] S230. Acquire multiple pieces of related information associated with the target event from multiple data sources according to the event triplet of the target event.
[0094] Specifically, the associated information is associated with the information about the target event. In other words, the event mentioned in the associated information is associated with the target event.
[0095] In some embodiments, S230 may include: S2321, obtaining from multiple data sources a plurality of related information whose publishing time is within a second preset time period, belongs to the target field, and is associated with the event subject and event information of the target event, wherein the second preset time period is determined by the event occurrence time of the target event.
[0096] Specifically, there are many ways to determine the second preset time period, which are described below with reference to typical examples, but are not limited thereto.
[0097] In one example, the second preset time period can be determined by the event occurrence time of the target event and the second preset duration. For example, the event occurrence time of the target event can be used as the start time of the second preset time period, and the time obtained by extending the event occurrence time of the target event by the second preset duration is used as the end time of the second preset time period, wherein the specific value of the second preset duration can be set by those skilled in the art according to actual conditions and is not limited here. For example, if the event occurrence time of the target event is 7:00 a.m. on May 10, 2023, and the second preset duration is one day, then the second preset time period is from 7:00 a.m. on May 10, 2023 to 7:00 a.m. on May 11, 2023.
[0098] In another example, the day on which the target event occurs may be used as the second preset time period.
[0099] Specifically, there are many specific implementation methods for determining whether a piece of information is associated with the event subject and event information of a target event. Typical examples are described below, but are not limited to these.
[0100] In one example, determining whether a piece of information is associated with an event subject and event information of a target event may include: performing word segmentation and part-of-speech analysis on the information to obtain multiple word segments and parts-of-speech of the multiple word segments; extracting the event subject and event information from the multiple word segments according to the parts-of-speech of the multiple word segments; determining the similarity between the event subject extracted from the information and the event subject of the target event (referred to as the first similarity), and determining the similarity between the event information extracted from the information and the event information of the target event (referred to as the second similarity); weighting the first similarity and the second similarity to obtain a total similarity; if the total similarity is greater than a sixth preset threshold, determining that the information is associated with the event subject and event information of the target event. Among them, the specific value of the sixth preset threshold can be set by a person skilled in the art according to actual conditions and is not limited here.
[0101] In another example, determining whether a piece of information is associated with the event subject and event information of a target event may include: determining the similarity between the information and the cluster center of the information group corresponding to the target event, and if the similarity is greater than a seventh preset threshold, determining that the information is associated with the event subject and event information of the target event. The specific value of the seventh preset threshold can be set by those skilled in the art according to actual conditions and is not limited here.
[0102] S240: Obtain at least one related event according to the plurality of related information.
[0103] Specifically, the associated event is associated with the target event.
[0104] In some embodiments, S240 may include: clustering multiple related information to obtain at least one information group; determining that an event corresponding to the information group is a related event, or, if the number of related information included in the information group is greater than an eighth preset threshold, determining that the event corresponding to the information group is a related event. The specific value of the eighth preset threshold can be set by those skilled in the art according to actual conditions and is not limited here.
[0105] It should be noted that "clustering multiple related information to obtain at least one information group" is similar to S1112 described above and will not be repeated here.
[0106] S250: If the correlation between the associated event and the target event meets a preset condition and the associated event is a hot event, determine that the associated event is a key event.
[0107] Specifically, key events are events that are related to the target event and that attract widespread attention in society, participate in discussions, arouse public emotions, and trigger strong reactions.
[0108] Specifically, the specific method of "detecting whether a key event is a hot key event" is similar to the specific implementation method of "detecting whether a target event is a hot key event". Exemplarily, the information group corresponding to the key event includes at least one sub-information group, and different sub-information groups come from different data sources; wherein, the detection of whether a key event is a hot event includes: for each sub-information group, counting the index parameters of the sub-information group, and analyzing the probability value of the key event being a hot event according to the index parameters of the sub-information group, and obtaining the probability value corresponding to the sub-information group, wherein the index parameters include the number of readings, the number of likes, the number of comments, the number of reposts, and the number of collections; adding up the probability values corresponding to each sub-information group to obtain the total probability value; if the total probability value is greater than the second preset threshold value, the key event is determined to be a hot event. Optionally, analyzing the probability value of a key event being a hot event according to the indicator parameters of the sub-information group to obtain the probability value corresponding to the sub-information group includes: performing weighted summation on the likes, comments, reposts, and favorites of the sub-information group to obtain the sum value; performing logarithmic operation on the reading volume of the sub-information group and the sum value to obtain the probability value corresponding to the sub-information group, wherein the base of the logarithmic operation is determined by the reading volume of the sub-information group, and the true number of the logarithmic operation is determined by the sum value. However, it is not limited thereto.
[0109] Specifically, there are many specific implementation methods for detecting whether the correlation between the associated event and the target event meets the preset conditions. Typical examples are described below, but are not limited to these.
[0110] In some embodiments, detecting whether the correlation between the associated event and the target event satisfies a preset condition may include:
[0111] S251. For a related event, search for related information published by a related account from the information group corresponding to the related event, and compare the related information published by the related account with the event information of the target event to obtain similarity, wherein the related account is an account associated with the event subject of the target event.
[0112] Specifically, when the event subject is a person, the account associated with the event subject may include the account registered by the event subject on any application, etc. When the event subject is an enterprise, unit, group, etc., the account associated with the event subject may include the account registered by the event subject or the founder of the event subject on any application, etc., but it is not limited to this. For example, if the event subject is a car company, the associated account may include the account registered by the founder of the car company on any application, the official account registered by the car company on any application, etc.
[0113] Specifically, the comparison may be made by using a comparison method such as BERT, and the similarity may include semantic cosine similarity, etc., but is not limited thereto.
[0114] Specifically, there are many specific implementation methods for comparing the associated information released by the associated account with the event information of the target event, and typical examples are described below.
[0115] In some embodiments, comparing the associated information published by the associated account with the event information of the target event to obtain the similarity may include: comparing the entire associated information published by the associated account with the event information of the target event to obtain the similarity.
[0116] In other embodiments, comparing the associated information posted by the associated account with the event information of the target event to obtain the similarity may include: performing word segmentation and part-of-speech analysis on the associated information posted by the associated account to obtain multiple word segments of the information and the parts-of-speech of the multiple word segments; extracting event information from the multiple word segments according to the parts-of-speech of the multiple word segments; and determining the similarity between the event information extracted from the information and the event information of the target event.
[0117] S252: If the similarity is greater than a third preset threshold, determine that the relevance between the associated event and the target event meets a preset condition.
[0118] Specifically, the specific value of the third preset threshold can be set by those skilled in the art according to actual conditions and is not limited here. For example, the third preset threshold can be 0.9, but is not limited thereto.
[0119] In other embodiments, detecting whether the correlation between the associated event and the target event satisfies the preset condition may include: determining the similarity between the cluster center of the information group corresponding to the associated event and the cluster center of the information group corresponding to the target event, and if the similarity is greater than an eighth preset threshold, determining that the correlation between the associated event and the target event satisfies the preset condition. The specific value of the eighth preset threshold can be set by those skilled in the art according to actual conditions and is not limited here.
[0120] S260: extracting event triples of key events from information groups corresponding to the key events.
[0121] Specifically, S260 is similar to the “extracting the event triplet of the target event from the information group corresponding to the target event” described in the above S120, and will not be repeated here.
[0122] S270: Generate an event context according to the event triples of the target event and the event triples of the key event.
[0123] Specifically, the event triples of the target event and the event triples of the key event are sorted according to the chronology of the event occurrence to generate an event context.
[0124] For example, the event triple of the target event includes "20221001, a certain enterprise, a certain event occurs", and the event triples of the three key events corresponding to the target event include "20221002, a certain enterprise, official response", "20221005, a certain enterprise, statement 1", and "20221010, a certain enterprise, statement 2". The event context generated accordingly is as follows:
[0125] ○20221001An incident occurred in a certain enterprise
[0126] ○20221002 Official response from a company
[0127] ○ 20221005 A company issued a statement1
[0128] ○ 20221010 A company issued a statement 2
[0129] ○20221011 End
[0130] The disclosed embodiment can first determine the target event, and when the target event is a hot event, extract the event triple of the target event from the information group corresponding to the target event, and then track the key event corresponding to the target event based on the event triple of the target event. In this way, the target event and the key event can be discovered separately and successively, so that the amount of information processed when discovering the target event and the key event is small, which is conducive to reducing the machine load, and can automatically search and discover the key event corresponding to the target event without entering keywords, improving the generalization ability, and combining natural language processing in the process of discovering key events, so that the judgment of key events is more accurate. It can be seen that the disclosed embodiment optimizes the discovery process of target events and key events, making the overall event context generation mechanism more perfect.
[0131] The following is an explanation with a specific example.
[0132] Acquire multiple information belonging to the vehicle field and released within a first preset time period from multiple data sources; cluster the multiple information to obtain at least one information group, wherein the information group includes at least one sub-information group, and different sub-information groups are derived from different data sources; if the number of information contained in the information group is greater than the first preset threshold, determine that the event corresponding to the information group is a target event. Then, for each sub-information group corresponding to the target event, count the index parameters of the sub-information group, and analyze the probability value of the target event as a hot event according to the index parameters of the sub-information group to obtain the probability value corresponding to the sub-information group, wherein the index parameters include the number of readings, the number of likes, the number of comments, the number of reposts, and the number of collections; add the probability values corresponding to each sub-information group to obtain the total probability value; if the total probability value is greater than the second preset threshold, determine that the target event is a hot event. Then, if the target event is a hot event, extract the event triple of the target event from the information group corresponding to the target event, wherein the event triple includes the time of event occurrence, the event subject, and the event information. Then, a plurality of related information with a release time within a second preset time period, belonging to the vehicle field, and associated with the event subject and event information of the target event is obtained from a plurality of data sources, wherein the second preset time period is determined by the event occurrence time of the target event; at least one related event is obtained based on the plurality of related information; for the related event, the related information released by the related account is searched from the information group corresponding to the related event, and the related information released by the related account is compared with the event information of the target event to obtain the similarity, wherein the related account is an account associated with the event subject of the target event; if the similarity is greater than a third preset threshold, it is determined that the correlation between the related event and the target event meets the preset condition; if the correlation between the related event and the target event meets the preset condition and the related event is a hot event, it is determined that the related event is a key event; the event triple of the key event is extracted from the information group corresponding to the key event; and an event context is generated based on the event triple of the target event and the event triple of the key event.
[0133] The disclosed embodiment integrates multiple data sources, customizes the hot event detection method, and combines natural language processing technical solutions such as cluster analysis and semantic similarity to finally complete the event context mining and generation. The disclosed embodiment can be applied in the process of enterprise public opinion monitoring to capture network public opinion event information and grasp the network event public opinion orientation in real time in massive data, which helps car companies to respond to event public opinion in the first place, and is conducive to providing data decision support for car companies from the user's perspective, thereby assisting in the targeted and accurate optimization of automobiles.
[0134] Figure 3 FIG. 1 is a schematic diagram showing the structure of an event context generation device provided by an embodiment of the present disclosure. The event context generation device can be understood as the above-mentioned electronic device or a part of the functional modules in the above-mentioned electronic device. Figure 3As shown, the event context generating device 300 includes:
[0135] An acquisition module 310 is used to acquire a target event;
[0136] The extraction module 320 is used to extract the event triple of the target event from the information group corresponding to the target event if the target event is a hot event, wherein the event triple includes the event occurrence time, the event subject and the event information;
[0137] The generating module 330 is used to track the target event according to the event triple of the target event and generate an event context.
[0138] The disclosed embodiment can obtain the target event; if the target event is a hot event, the event triple of the target event is extracted from the information group corresponding to the target event, wherein the event triple includes the time of event occurrence, event subject, and event information; the target event is tracked according to the event triple of the target event to generate an event context. It can be seen that the above technical solution can generate the event context of the hot event, so that the user can understand the development process of the hot event by browsing the event context of the hot event, which is conducive to improving the user's experience of watching information.
[0139] In another embodiment of the present disclosure, the acquisition module 310 may include:
[0140] A first acquisition submodule is used to acquire, from multiple data sources, multiple pieces of information that are published within a first preset time period and belong to a target field;
[0141] A first clustering submodule, used for clustering a plurality of information to obtain at least one information group;
[0142] The first determination submodule is used to determine that the event corresponding to the information group is a target event if the amount of information included in the information group is greater than a first preset threshold.
[0143] In another embodiment of the present disclosure, the information group corresponding to the target event includes at least one sub-information group, and different sub-information groups are derived from different data sources;
[0144] The device further includes a first detection module, which is used to detect whether the target event is a hot event. The first detection module may include:
[0145] The statistics and analysis submodule is used to count the index parameters of each sub-information group, and analyze the probability value of the target event being a hot event according to the index parameters of the sub-information group to obtain the probability value corresponding to the sub-information group, wherein the index parameters include the number of readings, the number of likes, the number of comments, the number of reposts, and the number of collections;
[0146] The summing submodule is used to sum the probability values corresponding to each sub-information group to obtain a total probability value;
[0147] The second determination submodule is used to determine that the target event is a hot event if the total probability value is greater than a second preset threshold.
[0148] In another embodiment of the present disclosure, the statistics and analysis submodule may include an analysis unit for analyzing the probability value of the target event being a hot event according to the indicator parameters of the sub-information group to obtain the probability value corresponding to the sub-information group, wherein the analysis unit may include:
[0149] The summing sub-unit is used to perform weighted summing of the likes, comments, reposts, and favorites of the sub-information group to obtain a sum value;
[0150] The logarithmic operation subunit is used to perform logarithmic operation according to the reading amount and the sum value of the sub-information group to obtain the probability value corresponding to the sub-information group, wherein the base of the logarithmic operation is determined by the reading amount of the sub-information group, and the true number of the logarithmic operation is determined by the sum value.
[0151] In another embodiment of the present disclosure, the generation module 330 may include:
[0152] A first acquisition submodule is used to acquire a plurality of associated information associated with the target event from a plurality of data sources according to the event triple of the target event;
[0153] A second acquisition submodule is used to acquire at least one associated event according to the plurality of associated information;
[0154] A third determination submodule is used to determine that the associated event is a key event if the relevance between the associated event and the target event meets a preset condition and the associated event is a hot event;
[0155] An extraction submodule, used for extracting event triples of key events from information groups corresponding to key events;
[0156] The generation submodule is used to generate event context according to the event triples of the target event and the event triples of the key event.
[0157] In another embodiment of the present disclosure, the first acquisition submodule is specifically used to obtain from multiple data sources multiple related information whose publishing time is within a second preset time period, belongs to a target field, and is associated with an event subject and event information of a target event, wherein the second preset time period is determined by the event occurrence time of the target event.
[0158] In another embodiment of the present disclosure, the device may further include a second detection module, which is used to detect whether the correlation between the associated event and the target event meets a preset condition. The second detection module may include:
[0159] The search and comparison submodule is used to search for the associated information published by the associated account from the information group corresponding to the associated event, and compare the associated information published by the associated account with the event information of the target event to obtain the similarity, wherein the associated account is the account associated with the event subject of the target event;
[0160] The fourth determination submodule is used to determine whether the relevance between the associated event and the target event meets a preset condition if the similarity is greater than a third preset threshold.
[0161] The device provided in this embodiment can execute the method of any of the above embodiments, and its execution method and beneficial effects are similar, which will not be repeated here.
[0162] Figure 4 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown.
[0163] like Figure 4 As shown, the electronic device may include a controller 401 and a memory 402 storing computer program instructions.
[0164] Specifically, the controller 401 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
[0165] The memory 402 may include a large capacity memory for information or instructions. By way of example and not limitation, the memory 402 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 402 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 402 may be inside or outside the integrated gateway device. In a particular embodiment, the memory 402 is a non-volatile solid-state memory. In a particular embodiment, the memory 402 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (Electrically Erasable Programmable ROM, EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.
[0166] The controller 401 reads and executes the computer program instructions stored in the memory 402 to perform the steps of the event context generation method provided in the embodiment of the present disclosure.
[0167] In one example, the electronic device may further include a transceiver 403 and a bus 404. Figure 4 As shown, the controller 401, the memory 402 and the transceiver 403 are connected via a bus 404 and communicate with each other.
[0168] The bus 404 includes hardware, software, or both. For example, but not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a Memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 404 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.
[0169] The embodiment of the present disclosure further provides a computer-readable storage medium, which may store a computer program. When the computer program is executed by a processor, the processor implements the event context generation method provided by the embodiment of the present disclosure.
[0170] The above-mentioned storage medium may, for example, include a memory 402 of computer program instructions, and the above-mentioned instructions may be executed by the controller 401 of the electronic device to complete the event context generation method provided by the embodiment of the present disclosure. Optionally, the storage medium may be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium may be a ROM, a random access memory (Random Access Memory, RAM), a compact disc read-only memory (Compact Disc ROM, CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.
[0171] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device 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 device.
[0172] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for generating an event context, characterized in that: include: Get the target event; If the target event is a hot event, extracting the event triple of the target event from the information group corresponding to the target event, wherein the event triple includes the event occurrence time, event subject and event information; The target event is tracked according to the event triplet of the target event to generate an event context.
2. The method according to claim 1, characterized in that The obtaining of the target event includes: Acquire multiple information belonging to a target field and having a release time within a first preset time period from multiple data sources; Clustering the plurality of information to obtain at least one information group; If the amount of information included in the information group is greater than a first preset threshold, then the event corresponding to the information group is determined to be the target event.
3. The method according to claim 1, characterized in that The information group corresponding to the target event includes at least one sub-information group, and different sub-information groups are derived from different data sources; Wherein, detecting whether the target event is a hot event includes: For each of the sub-information groups, the index parameters of the sub-information group are counted, and the probability value of the target event being a hot event is analyzed according to the index parameters of the sub-information group to obtain the probability value corresponding to the sub-information group, wherein the index parameters include the number of readings, the number of likes, the number of comments, the number of reposts, and the number of collections; Adding the probability values corresponding to each of the sub-information groups to obtain a total probability value; If the total probability value is greater than a second preset threshold, the target event is determined to be a hot event.
4. The method according to claim 3, characterized in that The step of analyzing the probability value of the target event being a hot event according to the indicator parameters of the sub-information group to obtain the probability value corresponding to the sub-information group includes: Performing weighted summation on the number of likes, comments, reposts, and favorites of the sub-information group to obtain a sum value; A logarithmic operation is performed based on the reading amount of the sub-information group and the sum value to obtain a probability value corresponding to the sub-information group, wherein the base of the logarithmic operation is determined by the reading amount of the sub-information group and the true number of the logarithmic operation is determined by the sum value.
5. The method according to claim 1, characterized in that The step of tracking the target event according to the event triplet of the target event to generate an event context includes: According to the event triplet of the target event, a plurality of related information associated with the target event is acquired from a plurality of data sources; Acquire at least one associated event according to the plurality of associated information; If the correlation between the associated event and the target event meets a preset condition, and the associated event is a hot event, determining the associated event as a key event; Extracting the event triplet of the key event from the information group corresponding to the key event; The event context is generated according to the event triplet of the target event and the event triplet of the key event.
6. The method according to claim 5, characterized in that The acquiring, according to the event triplet of the target event, a plurality of associated information associated with the target event from a plurality of data sources comprises: Acquire from multiple data sources the multiple related information whose publishing time is within a second preset time period, belongs to a target field, and is associated with an event subject and event information of the target event, wherein the second preset time period is determined by the event occurrence time of the target event.
7. The method according to claim 5, characterized in that Detecting whether the correlation between the associated event and the target event meets a preset condition includes: For the associated event, searching for associated information published by an associated account from the information group corresponding to the associated event, and comparing the associated information published by the associated account with the event information of the target event to obtain a similarity, wherein the associated account is an account associated with the event subject of the target event; If the similarity is greater than a third preset threshold, it is determined that the relevance between the associated event and the target event meets a preset condition.
8. An event context generating device, characterized in that: include: An acquisition module is used to acquire target events; An extraction module, for extracting an event triple of the target event from an information group corresponding to the target event if the target event is a hot event, wherein the event triple includes an event occurrence time, an event subject, and event information; A generation module is used to track the target event according to the event triple of the target event to generate an event context.
9. An electronic device, characterized in that: include: processor; A memory for storing executable instructions; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the event context generation method described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the event context generation method described in any one of claims 1 to 7.