CN107402925B - Information pushing method and device - Google Patents
Information pushing method and device Download PDFInfo
- Publication number
- CN107402925B CN107402925B CN201610334269.2A CN201610334269A CN107402925B CN 107402925 B CN107402925 B CN 107402925B CN 201610334269 A CN201610334269 A CN 201610334269A CN 107402925 B CN107402925 B CN 107402925B
- Authority
- CN
- China
- Prior art keywords
- event
- event type
- target
- type
- current news
- 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.)
- Active
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
Abstract
The invention provides an information pushing method and device, which are characterized in that a target object corresponding to a current news event is determined according to a target event type to which the current news event belongs, influence parameters of the target event type on the target object are determined according to a historical news event corresponding to the target event type, and the target object and the influence parameters are pushed to a user. The invention acquires the objects such as products or services directly related to the current news event, and compared with the prior art, the invention can save the time consumed by analyzing the related objects from a large amount of data with complex relationships by common individuals. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
Description
Technical Field
The invention belongs to the field of data processing, and particularly relates to an information pushing method and device.
Background
In real life, part of social events may have a certain influence on the trend of some industries, for example, part of news events may have obvious positive or negative influence on products or services provided in some industries and present relatively stable rules for the products or services.
At present, a data analysis platform is established by utilizing free information on the internet to analyze the influence of a large event on the financial industry, such as a foreign Kensho service, aiming at professionals in investment institutions, the Kensho service displays the influence data of all historical related events on the financial industry to the professionals in a listed mode, and the professionals need to analyze according to experience and serve as an important basis for the institutions to make financial schemes.
In general, an ordinary person is often interested in only a part of products or services, and the ordinary person lacks the ability to find an association between specific products or services related to a current news event from listed influence data, and for such characteristics of the ordinary person, the analysis result of the existing data analysis method cannot meet the requirements of the ordinary person, and cannot provide effective help for the ordinary person.
Disclosure of Invention
The invention provides an information pushing method and device, which are used for solving the problems that the existing analysis result of a news event is a large amount of data with complex relationship, the requirement of a common individual cannot be met, and effective help cannot be provided for the common individual.
In order to achieve the above object, the present invention provides an information pushing method, including:
determining a target object corresponding to the current news event according to the target event type of the current news event;
determining an influence parameter of the target event type on the target object according to the historical news event corresponding to the target event type;
and pushing the target object and the influence parameters to a user.
In order to achieve the above object, the present invention provides an information pushing apparatus, comprising:
the object acquisition module is used for determining a target object corresponding to the current news event according to the target event type to which the current news event belongs;
the determining module is used for determining the influence parameters of the target event type on the target object according to the historical news event corresponding to the target event type;
and the pushing module is used for pushing the target object and the influence parameters to a user.
According to the information pushing method and device, the target object related to the current news event is determined according to the target event type to which the current news event belongs, the influence parameter of the target event type on the target object is determined according to the historical news event corresponding to the target event type, and the target object and the influence parameter are pushed to the user. The invention acquires the objects such as products or services directly related to the current news event, and compared with the prior art, the invention can save the time consumed by analyzing the related objects from a large amount of data with complex relationships by common individuals. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
Drawings
Fig. 1 is a flowchart illustrating an information pushing method according to a first embodiment of the present invention;
fig. 2 is a flowchart illustrating an information pushing method according to a second embodiment of the present invention;
fig. 3 is a flowchart illustrating an information pushing method according to a third embodiment of the present invention;
fig. 4 is a schematic flowchart of an information push method according to a fourth embodiment of the present invention;
fig. 5 is a schematic application diagram of a fourth information pushing method according to an embodiment of the present invention;
fig. 6 is a second schematic application diagram of an information pushing method according to a fourth embodiment of the present invention;
fig. 7 is a third schematic application diagram of an information pushing method according to a fourth embodiment of the present invention;
FIG. 8 is a fourth schematic view illustrating an application of the information push method according to the fourth embodiment of the present invention;
fig. 9 is a schematic structural diagram of an information pushing apparatus according to a fifth embodiment of the present invention;
fig. 10 is a schematic structural diagram of an information pushing apparatus according to a sixth embodiment of the present invention.
Detailed Description
The following describes in detail an information pushing method and apparatus provided by an embodiment of the present invention with reference to the accompanying drawings.
Example one
Fig. 1 is a schematic flow chart illustrating an information pushing method according to a first embodiment of the present invention. The information pushing method comprises the following steps:
s101, determining a target object corresponding to the current news event according to the target event type of the current news event.
In real life, not all news events have certain influence on the trend of the industry, a continuous rule is formed, and the news events need to be classified from a large number of historical news events through technical means such as data mining and historical review.
The basic event types that have been identified today include social news events and company announcement type news events. For example, the earthquake magnitude exceeds a certain level, the population density of the occurrence place exceeds a certain level, and the social news event can be regarded as a social news event, and the high management of increasing or decreasing the bulletins and the proportion of increasing or decreasing the bulletins exceeds a certain level, and the social news event can be regarded as a company bulletin type news event.
In this embodiment, the target event type to which the current news event belongs needs to be determined according to the content of the current news event. Specifically, the keywords of the current news event may be obtained, and the target event type of the current news event is determined according to matching between the keywords and the keywords of the target event type.
Further, according to the influence of a large number of historical news events on the industry, specific products or services in the industry influenced by different event types can be obtained, wherein the products or services can be used as objects, and the object corresponding to each event type can be obtained. After the target event type of the current news event is determined, a target object corresponding to the target event type, namely a target object related to the current news event, can be obtained according to the corresponding relation between the event type and the target object.
S102, determining an influence parameter of the target event type on the target object according to the historical news event corresponding to the target event type.
After the target event type is determined, the historical news events included in the target event type can be acquired, after all the historical news events are acquired, influence data on the target object after the historical news events occur are further acquired, influence on the target object after the news events of the target event type occur can be acquired by analyzing the influence data, and the influence includes the degree of influence on the target object and the time of the influence. For example, when the target object is an investment product, the influence of a news event under the target event type on the price trend of the investment product and the duration of the influence of the news event on the price trend of the investment product can be analyzed. And taking the influence of the target event type on the target object as an influence parameter.
S103, pushing the target object and the influence parameters to a user.
After determining the influence parameters of the historical news events included in the target event type on the target object, the target object and the influence parameters of the historical news events on the target object can be pushed to the user, so that the user can make an investment strategy according to the influence parameters on the target object. Optionally, after the target object is obtained, the target object may be pushed to the user, when the user tries to know the target object, the target object may be selected through operations such as clicking, and then the influence parameter of the historical news event on the target object is pushed to the user.
According to the information pushing method provided by the embodiment, the target object related to the current news event is determined according to the target event type to which the current news event belongs, the influence parameter of the target event type on the target object is determined according to the historical news event corresponding to the target event type, and the target object and the influence parameter are pushed to the user. The invention acquires the objects such as products or services directly related to the current news event, and compared with the prior art, the invention can save the time consumed by analyzing the related objects from a large amount of data with complex relationships by common individuals. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
Example two
Fig. 2 is a schematic flow chart illustrating an information pushing method according to a second embodiment of the present invention. On the basis of the first embodiment, the determining a target object corresponding to the current news event according to the target event type to which the current news event belongs includes the following steps:
s201, classifying and processing the historical news events through data mining.
In real life, not all news events have certain influence on the trend of the industry, a continuous rule is formed, and the news events need to be classified from a large number of historical news events through technical means such as data mining and historical review.
The basic event types that have been identified today include social news events and company announcement type news events. For example, the earthquake magnitude exceeds a certain level, the population density of the occurrence place exceeds a certain level, and the social news event can be regarded as a social news event, and the high management of increasing or decreasing the bulletins and the proportion of increasing or decreasing the bulletins exceeds a certain level, and the social news event can be regarded as a company bulletin type news event.
S202, obtaining basic event types, a basic type word bank corresponding to each basic event type, event types included by each basic event type and an event type word bank corresponding to each event type.
Specifically, in the data mining process, extracting keywords corresponding to each historical news event, then performing intelligent matching on the keywords, in the intelligent matching process, obtaining the similarity among the keywords to obtain all the keywords belonging to the same basic event type, further obtaining the superior-inferior relation among the keywords, and effectively sequencing all the keywords according to the obtained superior-inferior relation among the keywords to obtain a basic event type lexicon.
Similarly, in the process of classifying a large number of historical news events by data mining, the event types included in each basic event type and the event type lexicon corresponding to each event type can be obtained. Specifically, the keyword combination with the highest matching degree is used as an optimal keyword combination, the optimal keyword combination is used for intelligently matching event type word banks corresponding to all event types under the target basic event type to obtain an event type with the highest matching degree with the optimal keyword combination, and the event type is used as the target event type of the current news event.
And S203, establishing a time sequence of each event type.
And S204, performing history backtesting on the time sequence of each event type to obtain an object related to each event type, and constructing a mapping relation between the event types and the objects.
Event classification is carried out on a large number of historical news events in a data mining mode, after the event types included by each basic event type are obtained, the time sequence of each event type is established, then historical retesting is carried out on the time sequence of each event type, objects related to each event type are obtained, and the mapping relation between the event types and the objects is established.
And S205, acquiring the target object according to the mapping relation and the target event type.
After the target event type of the current news event is determined, a target object corresponding to the target event type, namely a target object related to the current news event, can be obtained according to the mapping relation between the event type and the target object.
In the embodiment, a basic event type and a corresponding basic type lexicon are established through data mining, the event type and the event type lexicon included in the basic event type are obtained through historical return test, the mapping relation between the event type and the object is obtained, the event type affiliated to the current news event is intelligently matched through the keywords of the current news event, and the target object related to the current news event is determined according to the mapping relation. In the embodiment, the objects such as products or services directly related to the current news event are acquired, and compared with the prior art, the time consumed for analyzing the related objects from a large amount of data with complex relationships can be saved for a common person. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
EXAMPLE III
Fig. 3 is a schematic flow chart illustrating an information push method according to a third embodiment of the present invention. On the basis of the above embodiment, before determining the target object corresponding to the current news event according to the target event type to which the current news event belongs, the method further includes the following steps:
s301, obtaining keywords of the current news event.
In this embodiment, feature information extraction is performed on the text content of the current news event, so as to extract a plurality of corresponding keywords from the current news event. For example, the text content of the current news event may be subjected to feature information extraction using a K-nearest neighbor (KNN) text classification algorithm, a feature selection/support vector machine text algorithm. For example, the current news event is "national institute of improvement and development committee sends out a notice in 12 months and 16 days, and the highest retail price of gasoline and diesel is adjusted up from 12 months and 17 days", and after keyword extraction is performed on the news event, the following keywords are obtained: national committee for modification, 12 months 16 days, 12 months 17 days, up-regulation, gasoline and diesel oil, retail price and the like.
S302, classifying the event type of the current news event according to the keywords to obtain the target event type to which the current news event belongs.
In this embodiment, different basic event types are preset, and a basic type lexicon is constructed for each basic event type. The setting of the basic event type and the basic type word bank corresponding to the basic event type are obtained by data mining of a large number of historical news events. Specifically, in the data mining process, extracting keywords corresponding to each historical news event, then performing intelligent matching on the keywords, in the intelligent matching process, obtaining the similarity among the keywords to obtain all the keywords belonging to the same basic event type, further obtaining the superior-inferior relation among the keywords, and effectively sequencing all the keywords according to the obtained superior-inferior relation among the keywords to obtain a basic event type lexicon.
Specifically, all the keywords acquired in S301 are combined arbitrarily to acquire combinations of all the keywords of the current news event, then each keyword combination is intelligently matched with the basic type lexicon corresponding to the basic event type, and the basic event type corresponding to the keyword combination with the highest matching degree is used as the target basic event type. In this embodiment, when the keyword combinations are intelligently matched with the event type lexicon, a hash algorithm is mainly used to calculate the score of each keyword combination in the event type lexicon, and the keyword combination with the highest score is used as the keyword combination with the highest matching degree.
Each base event type also includes a plurality of event types, for example, a social news event may include: industry planning, industry adjustment and optimization, major technological progress, sudden/hot events and the like. In this embodiment, a plurality of event types are set for each basic event type in advance, and in order to identify the event types to which different news events belong, a corresponding event type lexicon is set for each event type in this embodiment. After each basic event type is obtained, intelligently matching keywords in all historical news events in the basic event type, obtaining semantic similarity among the keywords in the intelligent matching process, obtaining all keywords belonging to the same event type, obtaining the superior-inferior relation among the keywords, effectively sequencing all the keywords according to the obtained superior-inferior relation among the keywords, and obtaining an event type lexicon of the event type.
After the basic event type of the current news event is obtained, a target event type corresponding to the affiliation of the current news event is also required to be determined. Specifically, the keyword combination with the highest matching degree is used as an optimal keyword combination, the optimal keyword combination is used for intelligently matching event type word banks corresponding to all event types under the target basic event type to obtain an event type with the highest matching degree with the optimal keyword combination, and the event type is used as the target event type of the current news event.
According to the information pushing method provided by the embodiment, the event type affiliated to the current news event is intelligently matched through the keywords of the current news event, and the target object related to the current news event can be determined according to the mapping relation between the event type and the object. In the embodiment, the objects such as products or services directly related to the current news event are acquired, and compared with the prior art, the time consumed for analyzing the related objects from a large amount of data with complex relationships can be saved for a common person. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
Example four
Fig. 4 is a schematic flow chart illustrating an information push method according to a fourth embodiment of the present invention. On the basis of the above embodiment, the determining, according to the historical news event corresponding to the target event type, the influence parameter of the target event type on the target object includes the following steps:
s401, historical news events included in the target event type are obtained in a traversing mode.
After the target event type of the current news event is determined, the historical news event included in the target event type can be acquired in a traversal mode.
S402, establishing a time sequence for the target event type.
Specifically, after all the historical events of the target event type are obtained in a traversal manner, a time sequence is established for the target event type, wherein the time sequence can embody the time of occurrence of the historical news events included in the target event type and the duration of the influence on the target object after the occurrence of the historical news events.
S403, performing history retest on all historical news events based on the time sequence of the target event type, and acquiring the influence data of all historical news events on the target object.
After the time sequence of the target event type is established, history backtesting can be performed on all historical news events of the target event type based on the time sequence so as to obtain the influence data of each historical news event on the target object.
And S404, determining an influence parameter based on the influence data.
And acquiring the income condition or the price trend of the target object on the time sequence according to the influence data of the target event type on the target object, thereby forming an influence parameter on the target object after the target event type occurs. Wherein, the influence parameter can reflect the influence on the price trend of the target object and the influence duration after the news event under the target event type occurs. The influence force parameters can be displayed through an influence trend graph, so that a user can more intuitively see the influence on a target object after the target event type occurs, and further, an investment suggestion can be objectively presented for the user.
Further, according to the influence parameter on the target object after the target event type occurs, the influence of the current news event on the target object, that is, the degree of influence of the current news event on the target object, including the influence interval and the influence duration on the target object, can be predicted. For example, when the target object is an investment product, the price trend of the current news event for the investment product may be predicted, and the duration of the influence of the current news event on the price trend of the investment product may be predicted. For example, after receiving the trend graph of the influence of the event type on the target object, the user can determine that the influenced target object has more than 5% excess income within 1-5 days after the current news event occurs.
In order to better understand the method for acquiring the news event related object provided by the embodiment, two scenes are taken for illustration.
Scene one: when the basic event type of the current news type is a social news event:
current news events: the national development and improvement committee sends out a notice at 12 months and 16 days, and the highest retail price of gasoline and diesel oil is adjusted up from 12 months and 17 days.
The type of the current news event is identified and categorized and a target object associated therewith is derived, in this example, the investment product.
The specific treatment process comprises the following steps:
through the feature extraction of the current news event, the following keywords are obtained: "national committee for modification, day 16 of 12 months, day 17 of 12 months, up-regulation, gasoline and diesel, retail price".
After the keywords are obtained, combining the keywords, and then obtaining the keyword combination with the highest matching degree by using an intelligent matching algorithm as follows: national development and transformation commission +12 months and 17 days + up regulation + gasoline and diesel oil + retail price, and the basic event type corresponding to the keyword combination with the highest matching degree is a social event.
Traversing the event type in the social event by the keyword combination with the highest matching degree, correspondingly obtaining the target event type with the highest matching degree, wherein the target event type is the adjustment and optimization of the petrochemical industry in the industry development planning, and according to the mapping relation between the event type and the target, the target object which can be obtained is the target object with the highest correlation with the petrochemical industry, and comprises the following steps: yongda group, China petroleum, China aviation photoelectricity, vast new growth mixed fund, emerging industry mixed fund of medium oil strategy, and Yinhua leading strategy mixed fund.
Furthermore, the influence data of the target event type on the target object is traversed, and a time sequence is established for the target event type. The time series may embody events of the occurrence of the historical news event included in the target event type and the duration of the impact on the target object after the occurrence of the historical news event. And obtaining the income condition or the price trend of the target object on the time sequence according to the influence data of the target event type on the target object, thereby forming an influence trend graph of the target event type on the target object, and further objectively presenting an investment suggestion for a user. The influence trend graph can reflect the influence on the price trend of the target object after the news event under the target event type occurs and the influence duration.
After the influence parameters of the target object and the target event type on the target object are acquired, the information of the influence parameters of the target object and the target event type on the target object can be pushed to a user and displayed on a display interface of a mobile terminal corresponding to the user. As shown in fig. 5, a description of the current news event and a recommended target object, i.e., an investment product, associated with the current news event are presented on a display interface.
When a user clicks a recommended investment product, an influence trend graph which is the influence on the investment product after the same type of events happen historically can be displayed. In this example, the investment product clicked by the user is Yongda group, and as shown in FIG. 6, the influence of the historical occurrence of the news event belonging to the same target event type as the current news event on the Yongda group is obtained, and optionally, a historical review may be included, in which there may be the number of historical occurrences of the news event belonging to the same target event type as the current news event, and a simple summary of the influence on the income of the Yongda group after the occurrence. In fig. 6, when a news event belonging to the target event type with the current news event appears in the influence trend graph, the news event can be marked according to the time sequence of the occurrence, and when the user clicks the mark, the specific influence of the news event on the Yongda group can be seen.
Scene two: when the basic event type of the current news event is a company announcement type event:
current news events: in 2015, 12 and 25 months, the eastern letters, Dahua shares and Shanghai green newly generated the increase of the high pipe/stockholder, and the confidence of the high pipe/stockholder for the future development of the company is shown.
The type of the current news event is identified and categorized and a target object associated therewith is derived, in this example, the investment product.
The specific treatment process comprises the following steps:
through the feature extraction of the current news event, the following keywords are obtained: "12/25 th/2015, eastern letters, Dahua shares, Shanghai green New, Gaoguan/stockholder, Zengshon".
After the keywords are obtained, combining the keywords, and then obtaining the keyword combination with the highest matching degree by using an intelligent matching algorithm as follows: 25 days 12 and 25 months 2015 + eastern national letter + Dahua share + Shanghai green New + high pipe/stockholder + increase hold, and the basic event type corresponding to the keyword combination with the highest matching degree is a company announcement type event.
Traversing the event type in the social event by the keyword combination with the highest matching degree, and correspondingly obtaining the target event type with the highest matching degree, wherein the target event type is as follows: high management and support are increased, and according to the mapping relation between the event type and the object, the obtained target object comprises: eastern letters, dahua thigh, shanghai green novi, and huatai bery value increase mixed funds.
Furthermore, historical data of the target object influenced by the target event type is traversed, a time sequence is established for the target event type, the time sequence can embody the events of the historical news events included in the target event type and the duration of the influence on the target object after the historical news events occur. Further, according to historical data of the target object influenced by the target event type, the income condition or the price trend of the target object in the time sequence is obtained, so that an influence trend graph of the target object after the target event type occurs is formed, and the investment suggestion can be objectively presented for the user. The influence trend graph can reflect the influence on the price trend of the target object after the target event type occurs and the influence duration.
After the influence of the target object and the target event type on the target object is acquired, the information of the influence of the target object and the target event type on the target object can be pushed to a user and displayed on a display interface of a mobile terminal corresponding to the user. As shown in fig. 7, a description of the current news event and a recommended target object, i.e., an investment product, related to the current news event are presented on a display interface.
When a user clicks a recommended investment product, an influence trend graph which is the influence on the investment product after the same type of events happen historically can be displayed.
In this example, the investment product clicked by the user may be influenced by the historical occurrence of the news event belonging to the same target event type as the current news event, as shown in fig. 8, and optionally may include a historical review, which may include a brief summary of the number of times the news event belonging to the same target event type as the current news event occurred in the history and the influence of the occurrence on the profit of the investment product.
Compared with the prior art, the information pushing method provided by the embodiment can save the time consumed by analyzing the related objects from a large amount of data with complex relationships by a common person. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
EXAMPLE five
Fig. 9 is a schematic structural diagram of an information pushing apparatus according to a fifth embodiment of the present invention. The information push device includes: the device comprises an object acquisition module 11, a determination module 12 and a push module 13.
The object obtaining module 11 is configured to determine, according to a type of a target event to which a current news event belongs, a target object corresponding to the current news event.
In real life, not all news events have certain influence on the trend of the industry, a continuous rule is formed, and the news events need to be classified from a large number of historical news events through technical means such as data mining and historical review.
The basic event types that have been identified today include social news events and company announcement type news events. For example, the earthquake magnitude exceeds a certain level, the population density of the occurrence place exceeds a certain level, and the social news event can be regarded as a social news event, and the high management of increasing or decreasing the bulletins and the proportion of increasing or decreasing the bulletins exceeds a certain level, and the social news event can be regarded as a company bulletin type news event.
In this embodiment, the object obtaining module 11 needs to determine the target event type to which the current news event belongs according to the content of the current news event. Specifically, the object obtaining module 11 may obtain a keyword of the current news event, and determine the target event type of the current news event by matching the keyword with the keyword of the target event type.
Further, the object obtaining module 11 may obtain specific products or services in industries affected by different event types according to the influence of a large number of historical news events on the industries, where the products or services may serve as objects and can obtain an object corresponding to each event type. After determining the target event type of the current news event, the object obtaining module 11 may obtain a target object corresponding to the target event type, that is, a target object related to the current news event, according to the correspondence between the event type and the object.
And the determining module 12 is configured to determine, according to the historical news event corresponding to the target event type, an influence parameter of the target event type on the target object.
After determining the target event type, the determining module 12 may obtain the historical news events included in the target event type. After obtaining all the historical news events, the determining module 12 obtains the influence data on the target object after the historical news events occur, and can obtain the influence parameters on the target object after the news events of the target event type occur by analyzing the influence data, wherein the influence parameters include the degree of influence on the target object and the time of the influence. For example, when the target object is an investment product, the determination module 12 may analyze the effect of a news event under the target event type on the price trend of the investment product, and the duration of the effect of the news event on the price trend of the investment product. And taking the influence of the target event type on the target object as an influence parameter.
And the pushing module 13 is configured to push the target object and the influence parameter to a user.
After the determining module 12 determines the influence parameters of the historical news events included in the target event type on the target object, the pushing module 13 may push the target object and the influence parameters of the historical news events on the target object to the user, so that the user may make an investment strategy according to the influence parameters on the target object. Optionally, after the target object is obtained, the target object may be pushed to the user, when the user tries to know the target object, the target object may be selected through operations such as clicking, and then the influence parameter of the historical news event on the target object is pushed to the user.
The information pushing device provided in this embodiment determines, according to the target event type to which the current news event belongs, the target object corresponding to the current news event object, determines, according to the historical news event corresponding to the target event type, the influence parameter of the target event type on the target object, and pushes the target object and the influence parameter to the user. In the embodiment, the objects such as products or services directly related to the current news event are acquired, and compared with the prior art, the time consumed for analyzing the related objects from a large amount of data with complex relationships can be saved for a common person. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
EXAMPLE six
Fig. 10 is a schematic structural diagram of an information pushing apparatus according to a sixth embodiment of the present invention. The news event related object acquisition device comprises: the system comprises an object acquisition module 11, a determination module 12, a push module 13, a keyword acquisition module 14 and a classification module 15.
An optional structural manner of the object obtaining module 11 in this embodiment includes:
and the data mining unit 111 is used for classifying and processing the historical news events through data mining.
The building unit 112 is configured to obtain the basic event types, a basic type lexicon corresponding to each basic event type, an event type included in each basic event type, and an event type lexicon corresponding to each event type.
A first establishing unit 113 for establishing a time series of each event type.
The first history backtesting unit 114 is configured to perform history backtesting on the time series of each event type to obtain an object related to each event type, and construct a mapping relationship between the event type and the object.
A first obtaining unit 115, configured to obtain the target object according to the mapping relationship and the target event type.
And a keyword obtaining module 14, configured to obtain a keyword of the current news event before the object obtaining module 11 obtains the target event type to which the current news event belongs to determine the target object corresponding to the current news event.
And the classifying module 15 is configured to classify the current news event according to the keyword to obtain the target event type.
An optional structural manner of the determination module 12 in this embodiment includes:
and the traversing unit 121 is configured to traverse to acquire the historical news events included in the target event type.
A second establishing unit 122, configured to establish a time sequence for the target event type.
The second history backtesting unit 123 is configured to perform history backtesting on all history news events based on the time sequence of the target event type, and acquire influence data of all history news events on the target object.
A second obtaining unit 124 for determining the influence parameter based on the influence data.
An optional structure manner of the classifying module 15 in this embodiment includes:
a combining unit 151 for combining all keywords of the current news event;
a first classifying unit 152, configured to classify the current news event base according to each keyword combination to obtain a target base event type of the current news event;
a setting unit 153, configured to use a keyword combination corresponding to the target basic event type as an optimal keyword combination;
the second classifying unit 154 classifies the event type of the current news event according to the optimal keyword combination to obtain the target event type.
The keyword obtaining module 14 is specifically configured to perform feature information extraction on the text content of the current news event to obtain the keyword of the current news event.
The first classifying unit 152 is specifically configured to perform intelligent matching on each keyword combination and a basic type lexicon corresponding to a basic event type, and use the basic event type corresponding to the keyword combination with the highest matching degree as a target basic event type.
A second classifying unit 154, configured to perform intelligent matching on the event type lexicon corresponding to all event types included in the target basic event type by using the optimal keyword combination, so as to obtain an event type with the highest matching degree with the optimal keyword combination;
the setting unit 153 is further configured to use the event type with the highest matching degree with the optimal keyword combination as the target event type of the current news event.
Compared with the prior art, the method and the device have the advantage that the time consumed for analyzing the related objects from a large amount of data with complex relationships can be saved for ordinary persons. Furthermore, as the common individual does not have professional experience, the accuracy of the result obtained after analyzing a large amount of data is poor, and the invention directly provides objects such as related products or services, and the like, the accuracy of analysis is improved, the requirements of the common individual can be better met, and more effective guidance or help is provided for the common individual to make an investment scheme.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (16)
1. An information pushing method, comprising:
determining a target object corresponding to the current news event according to the target event type of the current news event;
determining an influence parameter of the target event type on the target object according to the historical news event corresponding to the target event type;
and pushing the target object and the influence parameters to a user.
2. The method of claim 1, wherein the determining the target object corresponding to the current news event according to the target event type to which the current news event belongs comprises:
classifying and processing historical news events through data mining;
acquiring basic event types, a basic type word bank corresponding to each basic event type, event types included by each basic event type and an event type word bank corresponding to each event type;
establishing a time sequence of each event type;
performing historical backtesting on the time sequence of each event type to obtain an object related to each event type, and constructing a mapping relation between the event types and the objects;
and acquiring the target object according to the mapping relation and the target event type.
3. The method of claim 2, wherein before determining the target object corresponding to the current news event according to the target event type to which the current news event belongs, the method further comprises:
acquiring a keyword of a current news event;
and classifying the event type of the current news event according to the keyword so as to obtain the target event type to which the current news event belongs.
4. The method according to any one of claims 1 to 3, wherein the determining the influence parameter of the target event type on the target object according to the historical news event corresponding to the target event type comprises:
traversing and acquiring historical news events included in the target event type;
establishing a time sequence for the target event type;
performing history backtesting on all historical news events based on the time sequence of the target event type, and acquiring influence data of all historical news events on the target object;
determining the impact parameter based on the impact data.
5. The method of claim 3, wherein the event type classification of the current news event according to the keyword to obtain the target event type to which the news event belongs comprises:
combining all keywords of the current news event;
classifying the current news event base according to each keyword combination to obtain the target base event type of the current news event;
taking the keyword combination corresponding to the target basic event type as an optimal keyword combination;
and classifying the event type of the current news event according to the optimal keyword combination to obtain the target event type.
6. The method according to claim 3 or 5, wherein the obtaining of the keywords of the current news event comprises:
and extracting characteristic information of the text content of the current news event to obtain the keywords of the current news event.
7. The method of claim 5, wherein the classifying the current news event base according to each keyword combination to obtain the target base event type of the current news event comprises:
intelligently matching each keyword combination with a basic type word bank corresponding to a basic event type;
and taking the basic event type corresponding to the keyword combination with the highest matching degree as a target basic event type.
8. The method according to claim 5, wherein the event type classification of the current news event according to the optimal keyword combination to obtain the target event type comprises:
intelligently matching event type word banks corresponding to all event types included in the target basic event type by using the optimal keyword combination to obtain an event type with the highest matching degree with the optimal keyword combination;
and taking the event type with the highest matching degree with the optimal keyword combination as the target event type affiliated to the current news event.
9. An information pushing apparatus, comprising:
the object acquisition module is used for determining a target object corresponding to the current news event according to the target event type to which the current news event belongs;
the determining module is used for determining the influence parameters of the target event type on the target object according to the historical news event corresponding to the target event type;
and the pushing module is used for pushing the target object and the influence parameters to a user.
10. The apparatus of claim 9, wherein the object acquisition module comprises:
the data mining unit is used for classifying and processing the historical news events through data mining;
the building unit is used for acquiring basic event types, a basic type word bank corresponding to each basic event type, event types included by each basic event type and an event type word bank corresponding to each event type;
a first establishing unit, configured to establish a time series of each event type;
the first history retest unit is used for performing history retest on the time sequence of each event type to obtain an object related to each event type and constructing a mapping relation between the event types and the objects;
and the first acquisition unit is used for acquiring the target object according to the mapping relation and the target event type.
11. The apparatus of claim 10, further comprising:
the keyword acquisition module is used for acquiring keywords of the current news event before the object acquisition module acquires the target event type to which the current news event belongs so as to determine a target object corresponding to the current news event;
and the classification module is used for classifying the event type of the current news event according to the keyword so as to acquire the target event type.
12. The apparatus according to any one of claims 9-11, wherein the determining module comprises:
the traversing unit is used for traversing and acquiring the historical news events included in the target event type;
the second establishing unit is used for establishing a time sequence for the target event type;
the second history retest unit is used for performing history retest on all history news events based on the time sequence of the target event type and acquiring the influence data of all history news events on the target object;
a second obtaining unit for determining the influence parameter based on the influence data.
13. The apparatus of claim 11, wherein the classification module comprises:
the combination unit is used for combining all keywords of the current news event;
the first classification unit is used for classifying the current news event base according to each keyword combination so as to obtain the target base event type of the current news event;
the setting unit is used for taking the keyword combination corresponding to the target basic event type as the optimal keyword combination;
and the second classification unit is used for classifying the event type of the current news event according to the optimal keyword combination so as to acquire the target event type.
14. The apparatus according to claim 11 or 13, wherein the keyword obtaining module is specifically configured to perform feature information extraction on the body content of the current news event to obtain the keyword of the current news event.
15. The apparatus according to claim 13, wherein the first classifying unit is specifically configured to perform intelligent matching on each keyword combination and a basic type lexicon corresponding to a basic event type, and use a basic event type corresponding to a keyword combination with a highest matching degree as a target basic event type.
16. The apparatus according to claim 13, wherein the second classifying unit is specifically configured to perform intelligent matching on event type word banks corresponding to all event types included in the target basic event type by using the optimal keyword combination, and obtain an event type with a highest matching degree with the optimal keyword combination;
the setting unit is further configured to use the event type with the highest matching degree with the optimal keyword combination as the target event type to which the current news event belongs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610334269.2A CN107402925B (en) | 2016-05-19 | 2016-05-19 | Information pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610334269.2A CN107402925B (en) | 2016-05-19 | 2016-05-19 | Information pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107402925A CN107402925A (en) | 2017-11-28 |
| CN107402925B true CN107402925B (en) | 2020-11-20 |
Family
ID=60394014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610334269.2A Active CN107402925B (en) | 2016-05-19 | 2016-05-19 | Information pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107402925B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110209930A (en) * | 2019-04-30 | 2019-09-06 | 平安科技（深圳）有限公司 | The investee's analysis method and device of knowledge based map |
| CN111209400B (en) * | 2020-01-03 | 2023-06-16 | 北京明略软件系统有限公司 | Data analysis method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226554A (en) * | 2012-12-14 | 2013-07-31 | 西藏同信证券有限责任公司 | Automatic stock matching and classifying method and system based on news data |
| CN104573054A (en) * | 2015-01-21 | 2015-04-29 | 杭州朗和科技有限公司 | Information pushing method and equipment |
| CN104902292A (en) * | 2015-05-20 | 2015-09-09 | 无锡天脉聚源传媒科技有限公司 | Television report-based public opinion analysis method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050071217A1 (en) * | 2003-09-30 | 2005-03-31 | General Electric Company | Method, system and computer product for analyzing business risk using event information extracted from natural language sources |
-
2016
- 2016-05-19 CN CN201610334269.2A patent/CN107402925B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226554A (en) * | 2012-12-14 | 2013-07-31 | 西藏同信证券有限责任公司 | Automatic stock matching and classifying method and system based on news data |
| CN104573054A (en) * | 2015-01-21 | 2015-04-29 | 杭州朗和科技有限公司 | Information pushing method and equipment |
| CN104902292A (en) * | 2015-05-20 | 2015-09-09 | 无锡天脉聚源传媒科技有限公司 | Television report-based public opinion analysis method and system |
Non-Patent Citations (1)
| Title |
|---|
| 基于灰色关联分析的中文新闻事件关联性识别;刘盼盼 等;《计算机应用》;20160210;第408-413页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107402925A (en) | 2017-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070391B (en) | Data processing method and device, computer readable medium and electronic equipment | |
| US20180270616A1 (en) | Method and apparatus for identifying types of user geographical locations | |
| US20140172415A1 (en) | Apparatus, system, and method of providing sentiment analysis result based on text | |
| CN109345417B (en) | Online assessment method and terminal equipment for business personnel based on identity authentication | |
| CN112256762B (en) | Enterprise portrait method, system, equipment and medium based on industrial map | |
| CN109102157A (en) | A kind of bank's work order worksheet processing method and system based on deep learning | |
| CN113627566A (en) | Early warning method and device for phishing and computer equipment | |
| CN107402925B (en) | Information pushing method and device | |
| CN113282754A (en) | Public opinion detection method, device, equipment and storage medium for news events | |
| CN111241153A (en) | Enterprise natural person entity comprehensive judgment alignment method and system | |
| CN111221873A (en) | Inter-enterprise homonym identification method and system based on associated network | |
| CN115396389A (en) | Internet of things information technology customer service system based on emotion energy perception | |
| CN116401601B (en) | Power failure sensitive user handling method based on logistic regression model | |
| CN111310961A (en) | Data prediction method, data prediction device, electronic equipment and computer readable storage medium | |
| EP3493082A1 (en) | A method of exploring databases of time-stamped data in order to discover dependencies between the data and predict future trends | |
| CN114817518A (en) | License handling method, system and medium based on big data archive identification | |
| CN111291042B (en) | Power data processing system and method for power supply service | |
| CN113392920A (en) | Method, apparatus, device, medium, and program product for generating cheating prediction model | |
| CN114529383A (en) | Method and system for realizing tax payment tracking and tax loss early warning | |
| CN114090601A (en) | Data screening method, device, equipment and storage medium | |
| KR102078541B1 (en) | Issue interest based news value evaluation apparatus and method, storage media storing the same | |
| CN114092230A (en) | Data processing method and device, electronic equipment and computer readable medium | |
| Mullins | Retrospective analysis of technology forecasting: In-scope extension | |
| Panteladis et al. | Fragmented integration and business cycle synchronization in the Greek regions | |
| CN113837518B (en) | Personnel processing capacity describing method based on historical processing data |
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 | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1247321 Country of ref document: HK |
|
| TA01 | Transfer of patent application right |
Effective date of registration: 20201014 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20201014 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |