CN111026965A - Hot topic tracing method and device based on knowledge graph - Google Patents

Hot topic tracing method and device based on knowledge graph Download PDF

Info

Publication number
CN111026965A
CN111026965A CN201911233131.3A CN201911233131A CN111026965A CN 111026965 A CN111026965 A CN 111026965A CN 201911233131 A CN201911233131 A CN 201911233131A CN 111026965 A CN111026965 A CN 111026965A
Authority
CN
China
Prior art keywords
information
event
price
hot topic
hot
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911233131.3A
Other languages
Chinese (zh)
Other versions
CN111026965B (en
Inventor
金业
闫佳丽
崔华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bank of China Ltd
Original Assignee
Bank of China Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Bank of China Ltd filed Critical Bank of China Ltd
Priority to CN201911233131.3A priority Critical patent/CN111026965B/en
Publication of CN111026965A publication Critical patent/CN111026965A/en
Application granted granted Critical
Publication of CN111026965B publication Critical patent/CN111026965B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Computational Linguistics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a hot topic tracing method and a device based on a knowledge graph, wherein the method comprises the following steps: acquiring event information of one or more information events related to a to-be-traced hot topic, wherein the to-be-traced hot topic is related to the one or more information events, and each information event is related to one or more information; according to event information of each information event related to the hot topic to be traced, one or more historical information events related to the hot topic to be traced are searched from a pre-constructed knowledge graph, wherein the knowledge graph comprises: the incidence relation between each hot topic and each information event; collecting information related to each information event and historical information events related to the hot topics to be traced; and outputting information associated with each information event and historical information events associated with the hot topics to be traced. According to the method and the system, based on each information event associated with the hot topic to be traced, historical similar information events associated with the hot topic to be traced can be searched.

Description

Hot topic tracing method and device based on knowledge graph
Technical Field
The invention relates to the field of internet, in particular to a hot topic tracing method and device based on a knowledge graph.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
With the rapid development of the internet, the information quantity acquired by a user through the internet shows explosive growth, in order to improve the reading efficiency of information content and rapidly extract effective information, one or more hot topics can be constructed according to the information in a certain time period, so that the user can rapidly trace the information associated with the hot topics concerned by the user. However, with the accumulation of information data of historical hot topics, for a certain hot topic, besides information of one or more information events directly related to the hot topic when the hot topic is constructed, some information events in the historical accumulated information data may be related to the hot topic, but since the occurrence time of the information events is earlier and the information events are already related to the historical hot topic, there is an urgent need for a method capable of searching historical similar information events related to the hot topic to be traced back so as to trace back more comprehensive information.
At present, the conventional information platform realizes the tracing of historical information in a way of searching for keywords, the information searched in this way is massive and cannot be quickly traced to the information heat trend of a certain hot topic, and because the way can only search for the information content containing the search keywords, some information contents which do not contain the search keywords but are related to the hot topic are not searched, so that the information traced to the certain hot topic is not comprehensive.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the invention provides a hot topic tracing method based on a knowledge graph, which is used for solving the technical problems that the existing information platform realizes the tracing of historical information in a way of searching keywords, so that the efficiency of tracing the information of a certain hot topic is low, and the information is incomplete, and comprises the following steps: acquiring event information of one or more information events related to a to-be-traced hot topic, wherein the to-be-traced hot topic is related to the one or more information events, and each information event is related to one or more information; according to event information of each information event related to the hot topic to be traced, one or more historical information events related to the hot topic to be traced are searched from a pre-constructed knowledge graph, wherein the knowledge graph comprises: the incidence relation between each hot topic and each information event; collecting information related to each information event and historical information events related to the hot topics to be traced; and outputting information associated with each information event and historical information events associated with the hot topics to be traced.
The embodiment of the invention also provides a hot topic tracing device based on the knowledge graph, which is used for solving the technical problems that the existing information platform realizes the tracing of historical information by searching keywords, so that the efficiency of tracing the information of a certain hot topic is low, and the information is incomplete, and the device comprises: the system comprises a hot topic information query module, a hot topic information query module and a hot topic tracking module, wherein the hot topic to be tracked is associated with one or more information events, and each information event is associated with one or more information; the knowledge graph searching module is used for searching one or more historical information events related to the hot topic to be traced from a pre-constructed knowledge graph according to the event information of each information event related to the hot topic to be traced, wherein the knowledge graph comprises: the incidence relation between each hot topic and each information event; the hot topic information acquisition module is used for acquiring information associated with each information event and historical information event associated with the hot topic to be traced; and the hot topic information output module is used for outputting information associated with each information event and historical information event associated with the hot topic to be traced.
The embodiment of the invention also provides computer equipment for solving the technical problems that the existing information platform realizes the tracing of historical information in a way of searching keywords, so that the efficiency of tracing the information of a certain hot topic is low, and the information is incomplete.
The embodiment of the invention also provides a computer readable storage medium, which is used for solving the technical problems that the existing information platform realizes the tracing of historical information by searching keywords, so that the efficiency of tracing the information of a certain hot topic is low, and the information is incomplete.
In the embodiment of the invention, after the event information of one or more information events associated with the hot topic to be traced is acquired, one or more historical information events associated with the hot topic to be traced are searched from a pre-constructed knowledge map according to the event information of each information event associated with the hot topic to be traced, the information associated with each information event and historical information event associated with the hot topic to be traced is acquired, and the information associated with each information event and historical information event associated with the hot topic to be traced is output.
According to the embodiment of the invention, based on each information event associated with the hot topic to be traced, the history similar information event associated with the hot topic to be traced is searched, so that the efficiency of tracing the history information can be greatly improved, and the information of the hot topic to be traced can be more comprehensively displayed.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
fig. 1 is a flowchart of a hot topic tracing method based on a knowledge graph according to an embodiment of the present invention;
FIG. 2 is a schematic illustration of a knowledge-graph provided in an embodiment of the present invention;
fig. 3 is a schematic diagram of an optional hot topic tracing interface provided in an embodiment of the present invention;
fig. 4 is a schematic diagram of a preferred hot topic tracing interface provided in an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a first timestamp and a second timestamp of an information event falling within a product price period according to an embodiment of the present invention;
FIG. 6 is a schematic diagram illustrating that neither the first timestamp nor the second timestamp corresponds to an information event falls within a product price period according to an embodiment of the present invention;
FIG. 7 is a diagram illustrating an information event according to an embodiment of the present invention, wherein a first timestamp falls within a product price time period and a second timestamp does not fall within the product price time period;
FIG. 8 is a diagram illustrating an information event according to an embodiment of the present invention, wherein a first timestamp does not fall within a product price time period and a second timestamp falls within the product price time period;
fig. 9 is a hot topic tracing device based on a knowledge graph provided in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
In the description of the present specification, the terms "comprising," "including," "having," "containing," and the like are used in an open-ended fashion, i.e., to mean including, but not limited to. Reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," etc., means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. The sequence of steps involved in the embodiments is for illustrative purposes to illustrate the implementation of the present application, and the sequence of steps is not limited and can be adjusted as needed.
The embodiment of the invention provides a hot topic tracing method based on a knowledge graph, fig. 1 is a flow chart of the hot topic tracing method based on the knowledge graph, as shown in fig. 1, the method can include the following steps:
s101, event information of one or more information events related to the hot topic to be traced is obtained, wherein the hot topic to be traced is related to the one or more information events, and each information event is related to one or more information.
It should be noted that the hot topic to be traced in the embodiment of the present invention may be a newly constructed hot topic, or may be any historical hot topic.
Before S101, the method for displaying the hot topic log information provided in the embodiment of the present invention may further implement construction of each hot topic by the following steps: collecting information of a plurality of information platforms; classifying the information into one or more information topics; the method comprises the steps of collecting heat information of each information topic in real time, and calculating a heat index of each information topic according to the heat information of each information topic, wherein the heat information of each information topic comprises the number of clicks, the number of praises or the number of comments of all information in each information topic; monitoring whether the heat index of each information topic is continuously higher than a first preset threshold value within a first preset time period; determining the information topic with the heat index continuously higher than a first preset threshold value in a first preset time period as the hot topic, creating a record document of the hot topic, recording the information of the hot topic until the heat index of the hot topic is continuously lower than a second preset threshold value in a second preset time period, and stopping recording the information of the hot topic.
According to the embodiment of the invention, the information of different information platforms is classified, the hot topics are extracted from the mass information, and the information of the hot topics is recorded, so that the efficiency of reading the information content by the user can be improved, the user can quickly know the influence trend of the concerned information topics, and for the investment user of the financial market tea product, the influence condition of the financial market product invested or invested with intent can be quickly known from the mass financial news information.
As different hot topics have different durations of heat change, some hot topics may last for one day or several days, and some hot topics may last for several months, even one year or several years, so that when constructing each hot topic, the heat index threshold (i.e., the second preset threshold) for determining whether each hot topic ends also needs to be set to a different value, as an optional implementation manner, the embodiment of the present invention may determine the heat index threshold at which each hot topic ends through the following steps: monitoring a duration of each hot topic; and determining a second preset threshold corresponding to each hot topic according to the duration time of each hot topic.
The information platform in the embodiment of the invention can be any platform for releasing information, including but not limited to microblogs, WeChat friend groups, Internet news and the like; as a preferred embodiment, the embodiment of the present invention is described by taking the example of publishing the information related to the financial market products, for example, the information platforms such as penbo, Wind, wale news, etc. will push a lot of financial news information every day, and the users usually read the financial news information to trend and predict the financial market products that they have invested.
As an optional implementation manner, the embodiment of the present invention may interface with each information platform through a unified information interface to collect information published on each information platform. In the prior art, a user checks news information through different information platforms, time and labor are consumed, and the information can be published on different information platforms on the same topic, so that a lot of time is wasted when the user checks the information repeatedly. The embodiment of the invention is connected with a plurality of information platforms through a uniform information interface, thereby not only providing more comprehensive information contents, but also playing a role in removing redundancy.
When the information is classified into one or more hot topics, various clustering algorithms can be adopted to classify the information from each information platform into different hot topics, and the conventional clustering methods are numerous, and are not described again in the embodiment of the invention.
Because the hot topic is associated with at least one information event, multiple pieces of information are associated (for example, after an event occurs, different information platforms issue one or more pieces of information to report the event), each piece of information has corresponding influence indexes which influence the change of the heat degree, such as the number of clicks, the number of praise or the number of comments, and the like, so that the heat degree index of each information event can be calculated by collecting the influence indexes, such as the number of clicks, the number of praise or the number of comments, and the like, of all the information associated with each information event, setting different weights for the different influence indexes, and performing weighted average. It should be noted that the influence indicators of each piece of information are changed in real time, and thus, the heat indicators of each information event are also changed in real time.
The embodiment of the invention determines the information topic of which the heat index is continuously higher than a certain threshold value in a period of time as the hot topic. The heat degree index of each information topic can be calculated in real time by monitoring the heat degree information of each information topic in real time, so that whether the heat degree index of each information topic is continuously higher than a certain threshold (namely a first preset threshold) within a preset time period can be monitored in real time. Generally, if the heat index of a hot topic is higher than a preset threshold value only at a certain moment and is lower than the preset threshold value at other moments, it indicates that the influence of the hot topic is not great, and thus the hot topic cannot be determined as the hot topic.
Optionally, the first preset time period for determining the start of a hot topic and the first preset threshold of the heat index in the embodiment of the present invention may be specifically set for different application scenarios, which is not limited in the embodiment of the present invention. For example, in one embodiment, for the same hot topic, different information sources (i.e., different information platforms) may be set with different heat index thresholds and weights, and the heat index thresholds of the information platforms are weighted and averaged to obtain the heat index threshold of the hot topic; in another embodiment, multiple heat index thresholds of multiple historical hot topics may be averaged, and the average value may be determined as the heat index threshold of the current hot topic.
S102, searching one or more historical information events related to the hot topic to be traced from a pre-constructed knowledge graph according to event information of each information event related to the hot topic to be traced, wherein the knowledge graph comprises: and the association relationship between each hot topic and each information event.
It should be noted that, in the embodiment of the present invention, a knowledge graph may be constructed in advance according to user information, product information of financial market products, topic information of hot topics, and events of information events associated with each hot topic, and fig. 2 is a schematic diagram of a knowledge graph provided in the embodiment of the present invention, as shown in fig. 2, in the knowledge graph, an entity includes: "user", "product", "information event", "hot topic"; wherein, the user and the product have an attention relationship; the "" user "" and the "" information event "" have a concern relationship; the user and the hot topic have an attention relationship; the product and the information event have an association relationship; the product and the hot topic have an association relationship; the "" information event "" and the "" hot topic "" have an association relationship.
As an alternative implementation, the above S102 may be implemented by the following steps: according to event information of each information event associated with the hot topic to be traced, one or more information events of which the similarity of each information event associated with the hot topic to be traced is higher than a preset threshold value are searched from a knowledge map, and the one or more information events of which the similarity is higher than the preset threshold value are determined as historical information events associated with the hot topic to be traced; according to the event information of each information event with the similarity higher than the preset threshold, searching the hot topics related to each information event with the similarity higher than the preset threshold from the knowledge graph, determining the searched hot topics as the similar hot topics of the hot topics to be traced, and determining each information event related to the similar hot topics as the historical information event related to the hot topics to be traced.
The event information of each information event at least comprises a title and summary content of each information event, so that when the similarity between each information event related to the hot topic to be traced and the historical information event is determined, the method can be realized by the following steps: comparing the titles of all the information events related to the hot topics to be traced with the titles of all the information events related to the hot topics in the knowledge graph, and calculating a first similarity; comparing the summary content of each information event associated with the hot topic to be traced with the summary content of each information event associated with each hot topic in the knowledge graph, and calculating a second similarity; and determining the similarity of each information event in the knowledge graph and each information event associated with the hot topic to be traced according to the first similarity and/or the second similarity.
The embodiment of the invention can extract the title and the summary content of each information event according to the event information of each information event, and can be realized by the following two implementation modes:
in a first optional implementation mode, heat information of each piece of information related to each information event is collected; calculating the heat index of each piece of information related to each information event according to the heat information of each piece of information related to each information event; and extracting the title and the summary content of each information event according to the information corresponding to the maximum heat index.
In a second optional implementation manner, counting the occurrence times of preset keywords in each piece of information associated with each information event; and extracting the title and the summary content of each information event according to the information with the maximum occurrence times of the preset keywords.
It should be noted that, when extracting the title and summary content of an information event according to the information content of a certain piece of information, the information title of the piece of information may be directly used as the title of the information event, and in addition, after performing natural language analysis (segmentation, clause, participle, part-of-speech tagging, etc.) on the information content of the piece of information, based on the corpus, the modifiers such as verb, pronoun, fixed language, and shape language in the information detail content are deleted, only the noun is retained, the first text content is obtained, and the first text content is used as the summary content of the information event; if the number of words of the first text content exceeds a preset number of words (for example, 160 words), only the contents of the first segment and the last segment in the first text content are retained to obtain a second text content, and the second text content is used as the summary content of the information event.
Preferably, because news information usually includes people, places and events due to the particularity of the news information, embodiments of the present invention may extract nouns such as names of people and places included in each piece of information based on a name information base and a place name information base constructed in advance to determine the title and summary content of each information event, thereby improving the accuracy of information extraction. Generally. The name or place name with the largest occurrence frequency in each piece of information is the person and place of the information event.
Optionally, after extracting and de-duplicating nouns such as names of people or names of places included in each piece of information, all the extracted words may be directly used as the title of each information event, or sentences including all the extracted words may be used as the title of each information event, or key information extracted from the sentences including all the extracted words may be used as the title of each information event; in addition, the paragraph with the largest number of the extracted words is used as the summary content of each information event, or the paragraph where the title sentence is located is used as the summary content of each information event, or the corresponding summary content is generated according to the extracted words.
S103, collecting information related to each information event and historical information events related to the hot topics to be traced.
It should be noted that, after one or more historical information events associated with the to-be-traced hot topic are found according to the event information of each information event associated with the to-be-traced hot topic based on the knowledge spectrogram, the information associated with each information event and the historical information event associated with the to-be-traced hot topic can be collected.
S104, outputting information related to each information event and historical information events related to the hot topics to be traced.
It should be noted that the hot topic tracing method provided by the embodiment of the present invention may further include the following steps: collecting information of a plurality of information platforms; and constructing one or more hot topics according to the information of the information platforms.
When one or more hot topics are constructed according to information of a plurality of information platforms, the method can be realized through the following steps: classifying information of a plurality of information platforms into one or more information topics; the method comprises the steps of collecting heat information of each information topic in real time, and calculating a heat index of each information topic according to the heat information of each information topic, wherein the heat information of each information topic comprises the number of clicks, the number of praise or the number of comments of all information related to each information topic; monitoring whether the heat index of each information topic is continuously higher than a first preset threshold value within a first preset time period; determining the information topic with the heat index continuously higher than a first preset threshold value in a first preset time period as the hot topic, creating a record document of the hot topic, recording the information of the hot topic until the heat index of the hot topic is continuously lower than a second preset threshold value in a second preset time period, and stopping recording the information of the hot topic.
It should be noted that each hot topic is associated with one or more information events, each information event is associated with one or more information, each information has an influence index such as its corresponding number of clicks, number of thumbs or number of comments, which can represent its popularity information, so that after the information of each information event associated with a certain hot topic is obtained, the popularity information of each information event in the hot topic, that is, the number of clicks, the number of thumbs or the number of comments of the information respectively associated with each information event, can be counted.
Optionally, the method for tracing hot topics provided by the embodiment of the present invention may further include the following steps: monitoring a duration of each hot topic; and determining a second preset threshold corresponding to each hot topic according to the duration time of each hot topic.
The hot topic tracing method provided by the embodiment of the invention can further comprise the following steps: counting the heat information of each information event according to the information related to each information event; calculating the heat index of each information event according to the heat information of each information event; according to the heat indexes of the hot topics to be traced and related to the information events, drawing a heat curve of the hot topics to be traced; and displaying the heat curve of the hot topic to be traced.
In order to facilitate the investment users of the financial market products to know the price change of the financial market products invested or invested with intention, which is influenced by the hot topic information event, the embodiment of the invention draws corresponding price curves according to the price data of each financial market product, and also outputs the price curves to the display interface of the hot topic, and can compare the price curves with the hot topic heat curves, so that the investment users of the financial market products can clearly know the influence of the hot topic on the price change of the financial market products.
Therefore, in an optional embodiment, the hot topic tracing method provided in the embodiment of the present invention may further include the following steps: inquiring one or more financial market products related to the hot topic to be traced from the knowledge map according to the hot topic to be traced and each information event related to the hot topic to be traced; collecting price data of one or more financial market products related to the hot topics to be traced; according to the price data of each financial market product associated with the hot topic to be traced, drawing a price curve of each financial market product associated with the hot topic to be traced in the continuous time period of the hot topic to be traced; and displaying the price curve of each financial market product related to the hot topic to be traced.
Preferably, a plurality of financial market products can be displayed on the interface, and the method for displaying the hot topic log information provided by the embodiment of the invention further comprises the following steps: receiving a second selection instruction, wherein the second selection instruction is used for selecting one or more financial market products displayed on the hot topic display interface; displaying, on the hot topic display interface, a price curve of the one or more financial market products selected by the second selection instruction.
Further, in order to quantify the influence of each information event in the hot topic on the price change of the financial market product and enable the user to more intuitively know the influence of the information event on the price wave of the financial market product concerned by the user, in an optional embodiment, the hot topic tracing method provided by the embodiment of the invention may further include the following steps: according to the event information of each information event, one or more financial market products related to each information event are searched from the knowledge graph; collecting price data of each financial market product related to each information event; calculating the price fluctuation rate of each information event on the influence of the price change of each financial market product according to the price data of each financial market product associated with each information event; outputting the price fluctuation rate of the influence of each information event on the price change of each financial market product.
Fig. 3 is a schematic view of a display interface of optional hot topic log information provided in an embodiment of the present invention, as shown in fig. 3, so that an investment user of a financial market product can quickly know an influence trend of a hot topic and an influence situation of the hot topic on price change of the financial market product. And after the user selects one or more financial market products concerned by the user, comparing and displaying the price curve of the financial market product selected by the user with the heat curve of the hot topic.
For the knowledge graph shown in fig. 2, after the "user 1" selects the "product 1", the "hot topic a" which affects the price of the "product 1" can be displayed, and the "information event 1", "information event 2", and "information event 3" directly associated with the "hot topic a" can be displayed on the display interface of the "hot topic a". Based on the knowledge graph, the 'information event 3' can be found to be also associated with the 'hot topic B' (namely, the 'hot topic B' has an information event with the similarity of the 'information event 3' of the 'hot topic A' being 100%), which indicates that the 'hot topic B' is a historical hot topic related to the 'hot topic A', and therefore the 'information event 6' associated with the 'hot topic B' is taken as a historical similar information event associated with the 'hot topic A'.
In addition, according to the event similarity, the information event 6 on the knowledge graph and the information event 6 of the hot topic A are found to have an association relation, and the hot topic D can be shown to be a historical hot topic related to the hot topic A; therefore, the "information event 6" is used as the history similar information event associated with the "hot topic A".
Further, according to topic similarity (topic similarity can be calculated by directly comparing titles or summary contents of two hot topics, and topic similarity of the two hot topics can also be obtained by weighted average according to event similarity of information events in the two hot topics), it can be found that the "hot topic C" on the knowledge graph is a historical hot topic related to the "hot topic a", and thus the "information event 7" of the "hot topic C" is also a historical similar information event related to the "hot topic a".
As shown in fig. 3, products such as "product 1", "product 2", "product 3", and the like displayed on the interface may be financial market products pushed according to the user history browsing information or hot topics, and hot topics such as "hot topic a" ("hot topic a" is directly associated with "product 1"), "hot topic B" ("hot topic B" is associated with the same information event as "hot topic a"), "hot topic C" ("hot topic C" has a certain topic similarity with "hot topic a"), "hot topic D" ("information event 6" of "hot topic D" has a certain topic similarity with the information event of "topic a") and the like displayed on the interface may be hot topics related to the user selection of the financial market product (for example, "product 1"); when a user selects a certain hot topic (for example, "hot topic a"), the log information of the hot topic can be displayed according to the time axis, and the information of each associated information event in the whole life cycle of the hot topic can be viewed by scrolling or sliding the time axis.
Optionally, for each information event, a title of each information event (e.g., "information event 1", "information event 2", "information event 3"), summary content of the information event, browsing volume, review volume, forwarding volume of the information event, information source platform (if there are more, one can click on "source" control to view each information platform), heat indicator value (the information event is associated with a weighted sum of all information heat information), and price fluctuation rate of price impact on the financial market product (e.g., product 1) price can be shown. The information events related to a certain hot topic can be sorted according to time or sorted according to heat. Alternatively, the ranking may be based on other indicators of the various information events (price volatility, forwarding amount, etc.).
For the historical similar information events of the "hot topic a", because the space on the interface is limited, the historical similar information events associated with the "hot topic a" can be presented according to a certain priority (the degree of association with the "hot topic a"), for example, "information event 4" and "information event 6" shown in fig. 3.
Fig. 4 is a schematic diagram of a display interface of preferred hot topic log information provided in an embodiment of the present invention, as shown in fig. 4, for the display of history similar information events, titles and summary information contents of the information events may also be displayed, the browsing volume, the comment volume, the forwarding volume, information source platforms (if there are a plurality of information events, a "source" control may be clicked to view the information platforms), a popularity indicator value (the information event is associated with a weighted sum of all information popularity information), and a price fluctuation rate affecting the price of a financial market product (for example, product 1).
Optionally, as shown in fig. 3 and 4, when each hot topic is displayed, some comment contents (for example, information hot development trend, price influence trend on financial market products, and the like) of the expert user on each hot topic can be displayed on the interface, so that a general user can quickly know a certain hot topic.
It should be noted that the hot-point curve above the interface is consistent with the information displayed below the interface, and the hot-point curve above the interface changes along with the rolling of the time axis below the interface; the displayed heat curve above the interface is moved, and the information below the interface is changed accordingly.
According to the embodiment of the invention, while the heat curve of a certain hot topic and the price curve of the related financial market product are displayed, the price fluctuation rate of the influence of each information event in the hot topic on the price change of the financial market product is displayed on the hot topic display interface, so that a user can not only quickly check the information of each information event in each hot topic, but also visually know the price fluctuation influence of each information event on the concerned financial market product.
As an alternative implementation manner, in the embodiment of the present invention, the price fluctuation rate of each information event on the influence of the price change of each financial market product is calculated by associating the price data of each financial market product with each information event, which may be specifically implemented by the following steps: acquiring a first time stamp and a second time stamp corresponding to each information event, wherein the first time stamp is a moment before the occurrence moment of each information event, and the second time stamp is a moment after the occurrence moment of each information event; determining a first price of the financial market product before each information event according to the first timestamp corresponding to each information event and the price time period of the financial market product for generating price data; determining a second price of the financial market product after each information event according to a second timestamp corresponding to each information event and a price time period of price data generated by the financial market product; determining a price fluctuation rate of each information event for the financial market product price change based on the first price and the second price.
Alternatively, the first price of the financial market product before each informational event occurs may be determined by: judging whether the first timestamp falls into a price time period for generating price data of the financial market product; if the first timestamp falls into the price time period of the financial market product for generating the price data, determining the average price of the financial market product between the first timestamp and the occurrence moment of each information event in the price time period of the financial market product for generating the price data or the price of the financial market product corresponding to the first timestamp as the first price of the financial market product before each information event occurs; if the first timestamp does not fall within the time period corresponding to the financial market product generated price data, determining the price of the financial market product corresponding to a price time period ending time before the first timestamp and adjacent to the first timestamp as the first price of the financial market product before each informational event occurs.
Alternatively, the second price of the financial market product after each informational event occurs may be determined by: judging whether the second timestamp falls into a price time period for generating price data of the financial market product; if the second timestamp falls into the price time period of the financial market product for generating the price data, determining the average price of the financial market product between the occurrence time of each information event and the second timestamp in the price time period of the financial market product for generating the price data or the price of the financial market product corresponding to the second timestamp as the second price of the financial market product after each information event occurs; if the second timestamp does not fall within the time period corresponding to the financial market product generated price data, determining the price of the financial market product corresponding to a price time period starting time after the second timestamp and adjacent to the second timestamp as the second price of the financial market product after each informational event occurs.
Since each hot topic usually comprises a plurality of information events, after the price fluctuation rate of each information event in the hot topic on the price change of each financial market product is determined, as a preferred implementation manner, the price fluctuation rate of the hot topic on the price change of each financial market product can be determined according to the price fluctuation rate of each information event in the hot topic on the price change of each financial market product.
In the embodiment of the present invention, it is assumed that the first timestamp and the second timestamp corresponding to an information event are Δ t respectively1And Δ t1The determination of the first price and the second price is specifically described below with reference to fig. 5 to 8:
as shown in fig. 5 or 7, the first time stamp Δ t1The price time period (i.e. the first price time period t) for which the financial market product generates price data falls1～t2) Determining the average price of the financial market products between a first time stamp and the occurrence time of each information event in the price time period of the financial market products generating the price data or the price of the financial market products corresponding to the first time stamp as the first price of the financial market products before each information event occurs; as shown in fig. 6 or 8, the first time stamp Δ t1And if the price does not fall into the time period corresponding to the price data generated by the financial market product, determining the price of the financial market product corresponding to the ending time of a price time period before the first time stamp and adjacent to the first time stamp as the first price of the financial market product before each information event.
As shown in fig. 5 or 8, the second time stamp Δ t2Price period falling into financial market product for generating price data (second time stamp Δ t in fig. 5)2Falls in a first price time period t1～t2Second time stamp Δ t in FIG. 82Falls into the second price period t3～t4) Determining the average price of the financial market products between the occurrence time of each information event and the second time stamp in the price time period of the financial market products for generating the price data or the price of the financial market products corresponding to the second time stamp as the second price of the financial market products after each information event occurs; as shown in fig. 6 or 7, the second time stamp Δ t2And if the price does not fall into the time period corresponding to the price data generated by the financial market product, determining the price of the financial market product corresponding to the starting time of a price time period which is after the second time stamp and is adjacent to the second time stamp as the second price of the financial market product after each information event.
It should be noted that the first time stamp Δ t in fig. 61And a second time stamp Δ t2None of the price time periods falls within the price time period for which the financial market products generate price data, and thus, the price of the financial market product corresponding to a price time period ending time before the first time stamp and adjacent to the first time stamp is determined as the first price of the financial market product before each information event occurs; and determining the price of the financial market product corresponding to a starting time of a price time period after and adjacent to the second time stamp as the second price of the financial market product after each informational event occurs.
For example, before and after an information event a occurs, the first price and the second price of the product a are P1 and P2, respectively, and the price fluctuation rate of the information event a on the product a is expressed as:
Figure BDA0002304109760000131
wherein Δ P represents the price fluctuation rate of the influence of the information event A on the product a.
In an optional embodiment, the hot topic tracing method provided in the embodiment of the present invention may further include the following steps: acquiring product information of one or more financial market products concerned by a user; according to the product information of each financial market product, one or more hot topics relevant to each financial market product are searched from the knowledge graph; displaying on the interface respective hot topics associated with each financial market product; receiving a topic selection instruction, wherein the topic selection instruction is used for selecting any one hot topic as a hot topic to be traced; displaying log information of the hot topic to be traced through a visual chart, wherein the log information at least comprises the following components: the method comprises the steps that titles, summary contents and popularity information of all information events related to hot topics to be traced, wherein the popularity information of each information event comprises the number of clicks, the number of praise or the number of comments of one or more pieces of information related to each information event.
According to the embodiment of the invention, the log information of each hot topic is displayed by displaying the visual chart, so that the information of the related information events in the life cycle of one hot topic can be visually displayed, a user can be helped to quickly trace all the information events related to one hot topic, the influence trend of the hot topic is known, and the reading efficiency of the information is improved.
It should be noted that the popularity information of each information event includes, but is not limited to, the total number of clicks, total number of praise, or total number of comments of all information associated with each information event; it should be noted that, when generating the log information of the topic, in addition to the above-mentioned title, summary content and heat information of each information event, other information of each information event may be combined, for example, each information event includes an information source of the information, a price fluctuation rate of each information event on the price of one or more financial market products, and the like.
Because the heat indexes of the hot topics change continuously along with time, in order to enable a user to visually know the heat change trend of the hot topics, a heat curve of the hot topics in a continuous time period can be drawn according to the heat indexes of the hot topics in the continuous time period, and the heat curve is output to a display interface of the hot topics. Therefore, after counting the heat information of each information event, as an optional embodiment, the method for displaying the log information of the hot topic provided by the embodiment of the present invention may further include the following steps: calculating the heat index of each information event according to the heat information of each information event; according to the heat indexes of all information events in the hot topics, drawing a heat curve of all the information events in the hot topics in the hot topic continuous time period; and displaying the heat curve of each information event in the hot topic duration time period on the interface.
Preferably, a plurality of hot topics can be displayed on the interface, and the method for displaying the log information of the hot topics provided by the embodiment of the present invention may further include the following steps: receiving a first selection instruction, wherein the first selection instruction is used for selecting one or more hot topics displayed on a hot topic display interface; and displaying the heat curve of the one or more hot topics selected by the first selection instruction on a hot topic display interface. After a user selects a certain hot topic, the heat curve of the hot topic can be displayed on the interface, preferably, the extreme point of the heat curve can be marked, and then the information of the hot event is displayed.
After determining the heat curve of each hot topic, as an optional implementation manner, the method for displaying the log information of the hot topics provided by the embodiment of the present invention may further include the following steps: one or more extreme points on the heat curve are obtained, and information corresponding to each extreme point is displayed on each extreme point of the heat curve, wherein the information corresponding to each extreme point is the information of a hot event. Through the implementation mode, the high tide point or the low tide point of the hot topic generation cycle can be displayed, and the information of the hot event corresponding to the high tide point or the low tide point is displayed.
When determining the extreme point of the heat curve, the method can be realized by the following steps: connecting the maximum value points on the heat curve to form an upper envelope line of the heat curve; connecting all minimum value points on the heat curve to form a lower envelope curve of the heat curve; averaging values corresponding to each moment on the upper envelope line and the lower envelope line to obtain a mean line of the heat curve; and determining one or more extreme points on the heat curve according to the upper envelope curve, the lower envelope curve or the mean line of the heat curve.
Alternatively, the maximum point of the upper envelope or the lower envelope may be used as the maximum point of the heat curve, and the minimum point of the upper envelope or the lower envelope may be used as the minimum point of the heat curve. Since the value of the upper envelope is generally larger than that of the lower envelope, it is preferable that the maximum point of the upper envelope is set as the maximum point of the heat curve and the minimum point of the lower envelope is set as the minimum point of the heat curve. In order to further improve the accuracy, the maximum point of the average line can be used as the maximum point of the heat curve, and the minimum point of the average line can be used as the minimum point of the heat curve.
Note that if the formed envelope is not smooth by connecting the respective extreme points (maximum points or minimum points) on the heat curve, the formed envelope is subjected to curve fitting until a smooth upper envelope and a smooth lower envelope are formed.
Further, after the hot topics are associated with the financial market products, the embodiment of the invention can push product information of some related financial market products to the user according to the hot topics concerned by the user, or push some related hot topic information to the user according to the financial market products invested by the user. Optionally, a product purchasing link may be added to the pushed product information or information, so that the user can quickly enter a corresponding purchasing interface while the user wants to purchase a certain financial market product.
As can be seen from the above, according to the hot topic tracing method based on the knowledge graph, after event information of one or more information events associated with a hot topic to be traced is obtained, one or more historical information events associated with the hot topic to be traced are searched from a pre-established knowledge graph according to the event information of each information event associated with the hot topic to be traced, information associated with each information event and historical information event associated with the hot topic to be traced is collected, and information associated with each information event and historical information event associated with the hot topic to be traced is output.
According to the hot topic tracing method based on the knowledge graph, the historical similar information events associated with the hot topics to be traced are searched based on the information events associated with the hot topics to be traced, so that the efficiency of tracing the historical information can be greatly improved, and the information of the hot topics to be traced can be displayed more comprehensively.
Based on the same inventive concept, the embodiment of the invention also provides a hot topic tracing device based on the knowledge graph, which is described in the following embodiment. Because the principle of solving the problems of the embodiment of the device is similar to the hot topic tracing method based on the knowledge graph, the implementation of the embodiment of the device can refer to the implementation of the method, and repeated parts are not described again.
Fig. 9 is a hot topic tracing apparatus based on a knowledge graph, as shown in fig. 9, the apparatus may include: the system comprises a hot topic information query module 901, a knowledge graph search module 902, a hot topic information acquisition module 903 and a hot topic information output module 904.
The hot topic information query module 901 is configured to obtain event information of one or more information events associated with a to-be-traced hot topic, where the to-be-traced hot topic is associated with the one or more information events, and each information event is associated with one or more pieces of information; the knowledge graph searching module 902 is configured to search, according to event information of each information event associated with a to-be-traced hot topic, one or more historical information events associated with the to-be-traced hot topic from a pre-constructed knowledge graph, where the knowledge graph includes: the incidence relation between each hot topic and each information event; the hot topic information acquisition module 903 is used for acquiring information associated with each information event and historical information events related to the hot topic to be traced; and a hot topic information output module 904, configured to output information associated with each information event and historical information events associated with the hot topic to be traced.
As can be seen from the above, in the hot topic tracing device based on the knowledge graph provided in the embodiment of the present invention, the hot topic information query module 901 queries event information of one or more information events associated with a hot topic to be traced; searching one or more historical information events associated with the hot topic to be traced from a pre-constructed knowledge graph through a knowledge graph searching module 902 according to the event information of each information event associated with the hot topic to be traced; the information related to each information event and historical information event related to the hot topic to be traced is collected through the hot topic information collection module 903, and the information related to each information event and historical information event related to the hot topic to be traced is output through the hot topic information output module 904.
According to the hot topic tracing device based on the knowledge graph, the historical similar information events related to the hot topics to be traced are searched based on the information events related to the hot topics to be traced, so that the efficiency of tracing the historical information can be greatly improved, and the information of the hot topics to be traced can be displayed more comprehensively.
In an optional embodiment, in the hot topic tracing apparatus provided in the embodiment of the present invention, the knowledge map searching module 902 is further configured to search, according to event information of each information event associated with the hot topic to be traced, one or more information events, of which the similarity of each information event associated with the hot topic to be traced is higher than a preset threshold, from the knowledge map, and determine the one or more information events of which the similarity is higher than the preset threshold as historical information events associated with the hot topic to be traced; according to the event information of each information event with the similarity higher than the preset threshold, searching the hot topics related to each information event with the similarity higher than the preset threshold from the knowledge graph, determining the searched hot topics as the similar hot topics of the hot topics to be traced, and determining each information event related to the similar hot topics as the historical information event related to the hot topics to be traced.
Optionally, the event information of each information event in the embodiment of the present invention at least includes a title and summary content of each information event, and thus, in an optional embodiment, the hot topic tracing apparatus provided in the embodiment of the present invention may further include: the information event similarity determining module 905 is used for comparing the titles of the information events related to the hot topics to be traced with the titles of the information events related to the hot topics in the knowledge graph, and calculating a first similarity; comparing the summary content of each information event associated with the hot topic to be traced with the summary content of each information event associated with each hot topic in the knowledge graph, and calculating a second similarity; and determining the similarity of each information event in the knowledge graph and each information event associated with the hot topic to be traced according to the first similarity and/or the second similarity.
In an optional embodiment, the device for tracing a hot topic provided in the embodiment of the present invention may further include: the information event information extracting module 906 is configured to extract a title and summary content of each information event according to the event information of each information event.
Based on the above embodiment, as a first optional implementation manner, the information event information extraction module 906 is further configured to collect heat information of each piece of information associated with each information event; calculating the heat index of each piece of information related to each information event according to the heat information of each piece of information related to each information event; and extracting the title and the summary content of each information event according to the information corresponding to the maximum heat index.
Based on the above embodiment, as a second optional implementation manner, the information event information extraction module 906 is further configured to count the occurrence frequency of a preset keyword in each piece of information associated with each information event; and extracting the title and the summary content of each information event according to the information with the maximum occurrence times of the preset keywords.
In an optional embodiment, the device for tracing a hot topic provided in the embodiment of the present invention may further include: the hot topic construction module 907 is used for collecting information of a plurality of information platforms; and constructing one or more hot topics according to the information of the information platforms.
Optionally, the hot topic constructing module 907 is further configured to classify information of multiple information platforms into one or more information topics; the method comprises the steps of collecting heat information of each information topic in real time, and calculating a heat index of each information topic according to the heat information of each information topic, wherein the heat information of each information topic comprises the number of clicks, the number of praise or the number of comments of all information related to each information topic; monitoring whether the heat index of each information topic is continuously higher than a first preset threshold value within a first preset time period; determining the information topic with the heat index continuously higher than a first preset threshold value in a first preset time period as the hot topic, creating a record document of the hot topic, recording the information of the hot topic until the heat index of the hot topic is continuously lower than a second preset threshold value in a second preset time period, and stopping recording the information of the hot topic.
In an optional embodiment, the device for tracing a hot topic provided in the embodiment of the present invention may further include: a hot topic monitoring module 908 for monitoring a duration period of each hot topic; and determining a second preset threshold corresponding to each hot topic according to the duration time of each hot topic.
In an optional embodiment, the device for tracing a hot topic provided in the embodiment of the present invention may further include: a heat curve drawing module 909 for counting the heat information of each information event according to the information related to each information event; calculating the heat index of each information event according to the heat information of each information event; according to the heat indexes of the hot topics to be traced and related to the information events, drawing a heat curve of the hot topics to be traced; and displaying the heat curve of the hot topic to be traced.
In an optional embodiment, the device for tracing a hot topic provided in the embodiment of the present invention may further include: the product price curve drawing module 910 is configured to query the knowledge graph for the hot topic to be traced associated with one or more financial market products according to the hot topic to be traced and each information event associated with the hot topic to be traced; collecting price data of one or more financial market products related to the hot topics to be traced; according to the price data of each financial market product associated with the hot topic to be traced, drawing a price curve of each financial market product associated with the hot topic to be traced in the continuous time period of the hot topic to be traced; and displaying the price curve of each financial market product related to the hot topic to be traced.
In an optional embodiment, the device for tracing a hot topic provided in the embodiment of the present invention may further include: a product price fluctuation rate calculation module 911, configured to search the knowledge graph for one or more financial market products associated with each information event according to the event information of each information event; collecting price data of each financial market product related to each information event; calculating the price fluctuation rate of each information event on the influence of the price change of each financial market product according to the price data of each financial market product associated with each information event; outputting the price fluctuation rate of the influence of each information event on the price change of each financial market product.
Optionally, the product price fluctuation rate calculation module 911 is further configured to obtain a first timestamp and a second timestamp corresponding to each information event, where the first timestamp is a time before the occurrence time of each information event, and the second timestamp is a time after the occurrence time of each information event; determining a first price of the financial market product before each information event according to the first timestamp corresponding to each information event and the price time period of the financial market product for generating price data; determining a second price of the financial market product after each information event according to a second timestamp corresponding to each information event and a price time period of price data generated by the financial market product; determining a price fluctuation rate of each information event for the financial market product price change based on the first price and the second price.
Optionally, the product price fluctuation rate calculation module 911 is further configured to determine whether the first timestamp falls within a price time period of the price data generated by the financial market product; if the first timestamp falls into the price time period of the financial market product for generating the price data, determining the average price of the financial market product between the first timestamp and the occurrence moment of each information event in the price time period of the financial market product for generating the price data or the price of the financial market product corresponding to the first timestamp as the first price of the financial market product before each information event occurs; if the first timestamp does not fall within the time period corresponding to the financial market product generated price data, determining the price of the financial market product corresponding to a price time period ending time before the first timestamp and adjacent to the first timestamp as the first price of the financial market product before each informational event occurs.
Optionally, the product price fluctuation rate calculation module 911 is further configured to determine whether the second timestamp falls within a price time period of the price data generated by the financial market product; if the second timestamp falls into the price time period of the financial market product for generating the price data, determining the average price of the financial market product between the occurrence time of each information event and the second timestamp in the price time period of the financial market product for generating the price data or the price of the financial market product corresponding to the second timestamp as the second price of the financial market product after each information event occurs; if the second timestamp does not fall within the time period corresponding to the financial market product generated price data, determining the price of the financial market product corresponding to a price time period starting time after the second timestamp and adjacent to the second timestamp as the second price of the financial market product after each informational event occurs.
In an optional embodiment, the device for tracing a hot topic provided in the embodiment of the present invention may further include: a hot topic log information display module 912, configured to obtain product information of one or more financial market products that a user pays attention to; according to the product information of each financial market product, one or more hot topics relevant to each financial market product are searched from the knowledge graph; displaying on the interface respective hot topics associated with each financial market product; receiving a topic selection instruction, wherein the topic selection instruction is used for selecting any one hot topic as a hot topic to be traced; displaying log information of the hot topic to be traced through a visual chart, wherein the log information at least comprises the following components: the method comprises the steps that titles, summary contents and popularity information of all information events related to hot topics to be traced, wherein the popularity information of each information event comprises the number of clicks, the number of praise or the number of comments of one or more pieces of information related to each information event.
The embodiment of the invention also provides computer equipment, which is used for solving the technical problems that the existing information platform realizes the tracing of historical information in a way of searching keywords, so that the efficiency of tracing the information of a certain hot topic is low, and the information is incomplete.
The embodiment of the invention also provides a computer readable storage medium, which is used for solving the technical problems that the existing information platform realizes the tracing of historical information by searching keywords, so that the efficiency of tracing the information of a certain hot topic is low, and the information is incomplete.
In summary, embodiments of the present invention provide a hot topic tracing method, an apparatus, a computer device, and a computer readable storage medium based on a knowledge graph, where after event information of one or more information events associated with a hot topic to be traced is obtained, one or more historical information events associated with the hot topic to be traced are searched from a pre-constructed knowledge graph according to event information of each information event associated with the hot topic to be traced, information associated with each information event and historical information event associated with the hot topic to be traced is collected, and information associated with each information event and historical information event associated with the hot topic to be traced is output.
According to the embodiment of the invention, based on each information event associated with the hot topic to be traced, the history similar information event associated with the hot topic to be traced is searched, so that the efficiency of tracing the history information can be greatly improved, and the information of the hot topic to be traced can be more comprehensively displayed.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (19)

1. A hot topic tracing method based on knowledge graph is characterized by comprising the following steps:
acquiring event information of one or more information events related to a to-be-traced hot topic, wherein the to-be-traced hot topic is related to the one or more information events, and each information event is related to one or more information;
according to event information of each information event related to the hot topic to be traced, one or more historical information events related to the hot topic to be traced are searched from a pre-established knowledge graph, wherein the knowledge graph comprises: the incidence relation between each hot topic and each information event;
collecting information related to each information event and historical information events related to the hot topic to be traced;
and outputting the information related to each information event and the historical information event related to the hot topic to be traced.
2. The method as claimed in claim 1, wherein the step of searching one or more historical information events associated with the hot topic to be traced from a pre-constructed knowledge graph according to the event information of each information event associated with the hot topic to be traced comprises:
according to the event information of each information event associated with the hot topic to be traced, one or more information events of which the similarity of each information event associated with the hot topic to be traced is higher than a preset threshold value are searched from a knowledge graph, and the one or more information events of which the similarity is higher than the preset threshold value are determined as historical information events associated with the hot topic to be traced;
according to the event information of each information event with the similarity higher than a preset threshold, searching hot topics related to each information event with the similarity higher than the preset threshold from a knowledge graph, determining the searched hot topics as similar hot topics of the hot topics to be traced, and determining each information event related to the similar hot topics as a historical information event related to the hot topics to be traced.
3. The method of claim 2, wherein the event information for each informational event includes at least a title and a summary content for each informational event, the method further comprising:
comparing the titles of the information events related to the hot topics to be traced with the titles of the information events related to the hot topics in the knowledge graph, and calculating a first similarity;
comparing the summary content of each information event associated with the to-be-traced hot topic with the summary content of each information event associated with each hot topic in the knowledge graph, and calculating a second similarity;
and determining the similarity of each information event in the knowledge graph and each information event associated with the hot topic to be traced according to the first similarity and/or the second similarity.
4. The method of claim 3, wherein the method further comprises:
and extracting the title and the summary content of each information event according to the event information of each information event.
5. The method of claim 4, wherein extracting a title and a summary content of each information event according to the event information of each information event comprises:
collecting heat information of each piece of information related to each information event;
calculating the heat index of each piece of information related to each information event according to the heat information of each piece of information related to each information event;
and extracting the title and the summary content of each information event according to the information corresponding to the maximum heat index.
6. The method of claim 4, wherein determining the title and summary content of each informational event based on the event information for each informational event comprises:
counting the occurrence times of preset keywords in each piece of information related to each information event;
and extracting the title and the summary content of each information event according to the information with the maximum occurrence times of the preset keywords.
7. The method of claim 1, wherein the method further comprises:
collecting information of a plurality of information platforms;
and constructing one or more hot topics according to the information of the information platforms.
8. The method of claim 7, wherein constructing one or more hot topics from information from a plurality of information platforms comprises:
classifying information of a plurality of information platforms into one or more information topics;
the method comprises the steps of collecting heat information of each information topic in real time, and calculating a heat index of each information topic according to the heat information of each information topic, wherein the heat information of each information topic comprises the number of clicks, the number of praise or the number of comments of all information related to each information topic;
monitoring whether the heat index of each information topic is continuously higher than a first preset threshold value within a first preset time period;
determining the information topic with the heat index continuously higher than a first preset threshold value in a first preset time period as the hot topic, creating a record document of the hot topic, recording the information of the hot topic until the heat index of the hot topic is continuously lower than a second preset threshold value in a second preset time period, and stopping recording the information of the hot topic.
9. The method of claim 8, wherein the method further comprises:
monitoring a duration of each hot topic;
and determining a second preset threshold corresponding to each hot topic according to the duration time of each hot topic.
10. The method of claim 1, wherein the method further comprises:
counting the heat information of each information event according to the information related to each information event;
calculating the heat index of each information event according to the heat information of each information event;
according to the hot indexes of the hot topics to be traced, which are related to the information events, drawing a hot curve of the hot topics to be traced;
and displaying the heat curve of the hot topic to be traced.
11. The method of claim 1, wherein the method further comprises:
inquiring one or more financial market products related to the hot topic to be traced from a knowledge graph according to the hot topic to be traced and each information event related to the hot topic to be traced;
collecting price data of one or more financial market products associated with the hot topic to be traced;
according to the price data of each financial market product related to the hot topic to be traced, drawing a price curve of each financial market product related to the hot topic to be traced in the continuous time period of the hot topic to be traced;
and displaying the price curve of each financial market product related to the hot topic to be traced.
12. The method of claim 1, wherein the method further comprises:
according to the event information of each information event, one or more financial market products related to each information event are searched from the knowledge graph;
collecting price data of each financial market product related to each information event;
calculating the price fluctuation rate of each information event on the influence of the price change of each financial market product according to the price data of each financial market product associated with each information event;
outputting the price fluctuation rate of the influence of each information event on the price change of each financial market product.
13. The method of claim 12, wherein calculating a rate of price volatility of each information event affecting price variations of a respective financial market product based on price data associated with the respective financial market product for each information event comprises:
acquiring a first time stamp and a second time stamp corresponding to each information event, wherein the first time stamp is a moment before the occurrence moment of each information event, and the second time stamp is a moment after the occurrence moment of each information event;
determining a first price of the financial market product before each information event according to a first timestamp corresponding to each information event and a price time period of price data generated by the financial market product;
determining a second price of the financial market product after each information event according to a second timestamp corresponding to each information event and a price time period of price data generated by the financial market product;
and determining the price fluctuation rate of each information event on the influence of the price change of the financial market product according to the first price and the second price.
14. The method of claim 13, wherein determining the first price of the financial market product prior to the occurrence of each informational event based on the first timestamp corresponding to each informational event and a price time period during which the financial market product generated price data comprises:
judging whether the first timestamp falls into a price time period of the financial market product for generating price data;
if the first timestamp falls into the price time period of the financial market product for generating price data, determining the average price of the financial market product between the first timestamp and the occurrence time of each information event in the price time period of the financial market product for generating price data or the price of the financial market product corresponding to the first timestamp as the first price of the financial market product before each information event occurs;
and if the first time stamp does not fall into the time period corresponding to the price data generated by the financial market product, determining the price of the financial market product corresponding to the price time period ending time before the first time stamp and adjacent to the first time stamp as the first price of the financial market product before each information event.
15. The method of claim 13, wherein determining the second price of the financial market product after the occurrence of each informational event based on the second timestamp corresponding to each informational event and a price time period for which the financial market product generated price data comprises:
judging whether the second timestamp falls into a price time period of the financial market product for generating price data;
if the second timestamp falls into the price time period of the financial market product for generating price data, determining the average price of the financial market product between the occurrence time of each information event in the price time period of the financial market product for generating price data and the second timestamp or the price of the financial market product corresponding to the second timestamp as the second price of the financial market product after each information event occurs;
if the second timestamp does not fall within the time period corresponding to the financial market product generated price data, determining the price of the financial market product corresponding to a price time period starting time after the second timestamp and adjacent to the second timestamp as the second price of the financial market product after each informational event occurs.
16. The method of any one of claims 1 to 15, further comprising:
acquiring product information of one or more financial market products concerned by a user;
according to the product information of each financial market product, one or more hot topics relevant to each financial market product are searched from the knowledge graph;
displaying on the interface respective hot topics associated with each financial market product;
receiving a topic selection instruction, wherein the topic selection instruction is used for selecting any one hot topic as the hot topic to be traced;
displaying log information of the topic to be traced back by a visual chart, wherein the log information at least comprises: the topic comprises a title, summary content and popularity information of each information event related to the to-be-traced hot topic, wherein the popularity information of each information event comprises the number of clicks, the number of praise or the number of comments of one or more information related to each information event.
17. A hot topic tracing device based on knowledge graph is characterized by comprising:
the system comprises a hot topic information query module, a hot topic information query module and a hot topic tracking module, wherein the hot topic to be tracked is associated with one or more information events, and each information event is associated with one or more information;
the knowledge map searching module is used for searching one or more historical information events related to the hot topic to be traced from a pre-constructed knowledge map according to the event information of each information event related to the hot topic to be traced;
the hot topic information acquisition module is used for acquiring information associated with each information event and historical information events related to the hot topic to be traced;
and the hot topic information output module is used for outputting information associated with each information event and historical information events associated with the hot topic to be traced.
18. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the method for tracing knowledge-graph based hot topics as claimed in any one of claims 1 to 16.
19. A computer-readable storage medium characterized in that the computer-readable storage medium stores a computer program for executing the knowledge-graph-based hot topic tracing method according to any one of claims 1 to 16.
CN201911233131.3A 2019-12-05 2019-12-05 Hot topic tracing method and device based on knowledge graph Active CN111026965B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911233131.3A CN111026965B (en) 2019-12-05 2019-12-05 Hot topic tracing method and device based on knowledge graph

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911233131.3A CN111026965B (en) 2019-12-05 2019-12-05 Hot topic tracing method and device based on knowledge graph

Publications (2)

Publication Number Publication Date
CN111026965A true CN111026965A (en) 2020-04-17
CN111026965B CN111026965B (en) 2023-11-21

Family

ID=70204309

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911233131.3A Active CN111026965B (en) 2019-12-05 2019-12-05 Hot topic tracing method and device based on knowledge graph

Country Status (1)

Country Link
CN (1) CN111026965B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111522967A (en) * 2020-04-27 2020-08-11 北京百度网讯科技有限公司 Knowledge graph construction method, device, equipment and storage medium
CN111753218A (en) * 2020-06-28 2020-10-09 中国银行股份有限公司 Hotspot knowledge determination method and related device
CN111881302A (en) * 2020-07-23 2020-11-03 民生科技有限责任公司 Bank public opinion analysis method and system based on knowledge graph
CN112559745A (en) * 2020-12-11 2021-03-26 科大讯飞股份有限公司 Method and related device for determining hot event
CN113722593A (en) * 2021-08-31 2021-11-30 北京百度网讯科技有限公司 Event data processing method and device, electronic equipment and medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070050273A1 (en) * 2005-08-23 2007-03-01 Logical Information Machines, Inc. System and method for presenting price movements before or following recurring historical events
CN109582783A (en) * 2018-10-26 2019-04-05 中国科学院自动化研究所 Hot topic detection method and device
CN110245298A (en) * 2019-06-18 2019-09-17 北京百度网讯科技有限公司 Method and apparatus for pushed information
CN110321482A (en) * 2019-06-11 2019-10-11 阿里巴巴集团控股有限公司 A kind of recommended method of information, device and equipment
CN110517149A (en) * 2019-08-29 2019-11-29 中国银行股份有限公司 The method and device of financial market machining information is pushed to user

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070050273A1 (en) * 2005-08-23 2007-03-01 Logical Information Machines, Inc. System and method for presenting price movements before or following recurring historical events
CN109582783A (en) * 2018-10-26 2019-04-05 中国科学院自动化研究所 Hot topic detection method and device
CN110321482A (en) * 2019-06-11 2019-10-11 阿里巴巴集团控股有限公司 A kind of recommended method of information, device and equipment
CN110245298A (en) * 2019-06-18 2019-09-17 北京百度网讯科技有限公司 Method and apparatus for pushed information
CN110517149A (en) * 2019-08-29 2019-11-29 中国银行股份有限公司 The method and device of financial market machining information is pushed to user

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111522967A (en) * 2020-04-27 2020-08-11 北京百度网讯科技有限公司 Knowledge graph construction method, device, equipment and storage medium
CN111522967B (en) * 2020-04-27 2023-09-15 北京百度网讯科技有限公司 Knowledge graph construction method, device, equipment and storage medium
CN111753218A (en) * 2020-06-28 2020-10-09 中国银行股份有限公司 Hotspot knowledge determination method and related device
CN111881302A (en) * 2020-07-23 2020-11-03 民生科技有限责任公司 Bank public opinion analysis method and system based on knowledge graph
CN111881302B (en) * 2020-07-23 2023-11-28 民生科技有限责任公司 Knowledge graph-based bank public opinion analysis method and system
CN112559745A (en) * 2020-12-11 2021-03-26 科大讯飞股份有限公司 Method and related device for determining hot event
CN112559745B (en) * 2020-12-11 2023-01-17 科大讯飞股份有限公司 Method and related device for determining hot event
CN113722593A (en) * 2021-08-31 2021-11-30 北京百度网讯科技有限公司 Event data processing method and device, electronic equipment and medium
CN113722593B (en) * 2021-08-31 2024-01-16 北京百度网讯科技有限公司 Event data processing method, device, electronic equipment and medium

Also Published As

Publication number Publication date
CN111026965B (en) 2023-11-21

Similar Documents

Publication Publication Date Title
CN111026965B (en) Hot topic tracing method and device based on knowledge graph
CN108829858B (en) Data query method and device and computer readable storage medium
Rastkar et al. Automatic summarization of bug reports
US8990241B2 (en) System and method for recommending queries related to trending topics based on a received query
CN101408886B (en) Selecting tags for a document by analyzing paragraphs of the document
CN105069102A (en) Information push method and apparatus
US20140172415A1 (en) Apparatus, system, and method of providing sentiment analysis result based on text
CN105138670A (en) Audio file label generation method and system
CN102831193A (en) Topic detecting device and topic detecting method based on distributed multistage cluster
CN105426514A (en) Personalized mobile APP recommendation method
US9245035B2 (en) Information processing system, information processing method, program, and non-transitory information storage medium
EP2577521A2 (en) Detection of junk in search result ranking
WO2022141803A1 (en) Method for automatically discovering hot keywords and hot news
CN105740353A (en) Calculation method and system for relevance degree of individual share and article
CN103020212A (en) Method and device for finding hot videos based on user query logs in real time
JP4896132B2 (en) Information retrieval method and apparatus reflecting information value
CN111523006A (en) Network public opinion tracking method for epidemic situation area
CN111046281B (en) Hot topic construction method and device
WO2022267325A1 (en) News popularity calculation method, device and storage medium
CN110929198B (en) Hot event display method and device
CN113642867A (en) Method and system for assessing risk
Ceroni et al. Justevents: A crowdsourced corpus for event validation with strict temporal constraints
CN111966899A (en) Search ranking method, system and computer readable storage medium
CN111026990B (en) Hot topic log information display method and device
CN114443930A (en) News public opinion intelligent monitoring and analyzing method, system and computer storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant