CN112364135B

CN112364135B - Object pushing method, device, equipment and storage medium based on multi-source data

Info

Publication number: CN112364135B
Application number: CN202011392838.1A
Authority: CN
Inventors: 关博睿; 毛才斐
Original assignee: Ping An Property and Casualty Insurance Company of China Ltd
Current assignee: Ping An Property and Casualty Insurance Company of China Ltd
Priority date: 2020-12-03
Filing date: 2020-12-03
Publication date: 2023-11-07
Anticipated expiration: 2040-12-03
Also published as: CN112364135A

Abstract

The invention relates to an intelligent recommendation technology and provides an object pushing method, device and equipment based on multi-source data and a storage medium. The method comprises the steps of obtaining a plurality of initial keywords by obtaining a text of a specific type in a preset time period, selecting a first preset number of target keywords based on frequency and preset weight, inputting the target keywords into a first model, inputting the target keywords into a second model when the first model outputs a first result, obtaining parameter information of a plurality of initial objects corresponding to the second result when the second model outputs a second result, screening a second preset number of target objects from the plurality of initial objects based on the parameter information and a plurality of preset dimensions, and pushing the target objects to a preset user group. The invention also relates to the technical field of blockchain, and the target keywords, parameter information and the like can be stored in a node of a blockchain.

Description

Object pushing method, device, equipment and storage medium based on multi-source data

Technical Field

The present invention relates to the field of intelligent recommendation technologies, and in particular, to an object pushing method, device, equipment and storage medium based on multi-source data.

Background

Currently, regarding the recommendation of an object, an evaluator usually evaluates the object subjectively according to some related attributes of the object and then recommends the object. For example, when an insurance company pushes a policy product to a user, a recommendation is made according to personal information, history policy and the like of the user after experience evaluation, but the accuracy of the recommendation result is low.

Although the technical schemes of automatic recommendation appear on the market, the schemes are usually based on a classification algorithm, and the technical problems of low accuracy, insufficient stability, high requirement on system performance and the like exist.

Disclosure of Invention

In view of the above, the present invention provides a method, an apparatus, a device and a storage medium for pushing objects based on multi-source data, which aims to solve the technical problem of low accuracy of pushing objects in the prior art.

In order to achieve the above object, the present invention provides an object pushing method based on multi-source data, the method comprising:

acquiring a text of a specific type from a preset database, preprocessing the text of the specific type to obtain a plurality of initial keywords, calculating the frequency of each initial keyword in the text of the specific type, and screening a first preset number of target keywords based on the frequency and a preset weight;

Inputting the target keyword into a first model, judging whether the output result of the first model is a first result, if so, inputting the target keyword into a second model, judging whether the output result of the second model is a second result, and if so, acquiring parameter information of a plurality of initial objects corresponding to the second result;

and screening a second preset number of target objects from the plurality of initial objects based on the parameter information and the preset dimensions of the plurality of initial objects, and pushing the target objects to a preset user group.

Preferably, the preprocessing the text of the specific type to obtain a plurality of keywords includes:

and executing word segmentation operation on the text of the specific type to obtain a plurality of segmented words, and extracting a plurality of initial keywords from the plurality of segmented words based on a TF-IDF algorithm.

Preferably, the selecting the first preset number of target keywords based on the frequency number and the preset weight includes:

and distributing associated weights to initial keywords with preset parts of speech in the plurality of initial keywords based on a preset distribution rule, calculating to obtain scores of the initial keywords based on the frequency and the weights, sequencing the scores from large to small, and selecting a first preset number of initial keywords as the target keywords.

Preferably, before the specific type of text is obtained from the preset database, the method further includes:

and acquiring a text of a specific type from a preset data source in real time, extracting structural information of the text of the specific type, and storing the structural information into the preset database.

Preferably, the method further comprises:

when the output result of the first model is judged not to be the first result, sending first prompt information to a preset user side;

and when the output result of the second model is judged not to be the second result, sending second prompt information to a preset user side.

Preferably, the second model is obtained by performing a clustering operation on a preset sample object set based on a clustering algorithm, and the specific clustering operation includes:

setting a density radius between sample objects and the number of sample objects with the smallest density radius, and based on the density radius and the number of sample objects with the smallest density, iteratively calculating to obtain a core sample object, a sample object with the reachable density and an edge sample object from all sample objects, obtaining the sample object with the reachable density of the core sample object, and updating a cluster corresponding to the core sample object by utilizing the sample object with the reachable density obtained by the iterative calculation.

Preferably, the pushing modes include a first pushing mode and a second pushing mode, and the first pushing mode includes: when a request for acquiring an object is received from a user in a preset user group, carrying out identity verification on the user, and pushing the target object to the user when the identity verification passes;

the second pushing mode includes: and acquiring the related information of the preset user group, wherein the related information comprises the IP address of each user, and pushing the target object to the preset user group according to the related information.

In order to achieve the above object, the present invention further provides an object pushing apparatus based on multi-source data, including:

and a pretreatment module: the method comprises the steps of obtaining a text of a specific type from a preset database, preprocessing the text of the specific type to obtain a plurality of initial keywords, calculating the frequency of each initial keyword in the text of the specific type, and screening a first preset number of target keywords based on the frequency and a preset weight;

and a judging module: the method comprises the steps of inputting the target keyword into a first model, judging whether an output result of the first model is a first result, if so, inputting the target keyword into a second model, judging whether an output result of the second model is a second result, and if so, acquiring parameter information of a plurality of initial objects corresponding to the second result;

And the pushing module is used for: and the target objects are used for screening a second preset number of target objects from the plurality of initial objects based on the parameter information and the preset dimensions of the plurality of initial objects, and pushing the target objects to a preset user group.

To achieve the above object, the present invention also provides an electronic device including:

at least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein,

the memory stores a program executable by the at least one processor to enable the at least one processor to perform any of the steps of the multi-source data based object pushing method as described above.

To achieve the above object, the present invention also provides a computer-readable storage medium storing a multi-source data-based object pushing program which, when executed by a processor, implements any of the steps of the multi-source data-based object pushing method described above.

According to the object pushing method, device, equipment and storage medium based on the multi-source data, the target keywords corresponding to the text of the specific type are obtained according to the text of the specific type, the target keywords are input into the model to obtain the parameter information of a plurality of initial objects, the target objects are screened out from the plurality of initial objects based on the parameter information and a plurality of preset dimensions, the target objects are pushed to the preset user group, and the accuracy of object pushing can be improved by combining the target keywords obtained by the text of the specific type.

Drawings

FIG. 1 is a flow chart illustrating a method for pushing objects based on multi-source data according to a preferred embodiment of the present invention;

FIG. 2 is a block diagram of an object pushing apparatus based on multi-source data according to a preferred embodiment of the present invention;

FIG. 3 is a schematic diagram of an electronic device according to a preferred embodiment of the present invention;

the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The invention provides an object pushing method based on multi-source data. Referring to fig. 1, a method flow diagram of an embodiment of a multi-source data-based object pushing method according to the present invention is shown. The method may be performed by an electronic device, which may be implemented in software and/or hardware, which may include, but is not limited to, smart phones, personal computers, notebook computers, tablet computers, portable wearable devices, and the like. The object pushing method based on the multi-source data comprises the following steps:

Step S10: acquiring a text of a specific type from a preset database, preprocessing the text of the specific type to obtain a plurality of initial keywords, calculating the frequency of each initial keyword in the text of the specific type, and screening a first preset number of target keywords based on the frequency and a preset weight.

In this embodiment, the present scheme is described by taking pushing of an insurance policy product in the field of insurance as an example, and it should be noted that the application scenario of the present scheme is not limited to this. The preset database may be an internal database of an insurance company or a third party database, the text of a specific type may be a corresponding text of webpage news of a preset type (for example, insurance class or social hotspot class) within a preset time period (for example, 3 days), a preprocessing operation is performed on the text of the specific type to obtain a plurality of initial keywords, the frequency of each keyword within the preset time period (for example, 3 days) is calculated and ranked from high to low, and a first preset number (for example, 10) of initial keywords are selected as target keywords, for example, the selected target keywords may be typhoons, mud-rock flows, storm and the like. Furthermore, a text library with a text source dimension and a text library with a time axis dimension can be constructed, so that a user can conveniently and quickly acquire required texts.

In one embodiment, the preprocessing the text of the specific type to obtain a plurality of keywords includes:

The text may be subjected to word segmentation operation by using a forward maximum matching algorithm or a reverse maximum matching algorithm to obtain a plurality of word segments, and the specific word segmentation mode is not limited herein. And extracting a plurality of initial keywords in a plurality of segmented words according to a TF-IDF algorithm, counting word frequencies of all words appearing in the text in advance, calculating an IDF value, and then calculating a TF value for each word of the text. Wherein tf= (number of occurrences of words in text)/(total number of words in text), the TF-ID value of the word is obtained by multiplying the IDF value by the TF value, and the TF-ID value can evaluate the importance of the word in the text, and a larger TF-ID value indicates a higher priority as a keyword. When TF-IDF calculation is performed, a TF-IDF value of a word is obtained by using word frequency (TF) and Inverse Document Frequency (IDF), and if the TF-IDF value is larger, the importance of the word to a text is higher, so that the word with the TF-IDF value arranged in front can be used as an initial keyword.

In one embodiment, the selecting the first preset number of target keywords based on the frequency and the preset weight includes:

Since the TF-IDF algorithm does not consider the degree of distinction of the text by the location factor of the feature words, the contribution to the degree of distinction is different when the keywords appear in different locations of the text. For example, the text of web page news is typically composed of a title, a body, and often the most important parts are the title, the body start, and the body end. The text is made up of sentences, usually the most important part of a sentence is the host-predicate, while the host and object are generally nouns and the predicate is generally a verb. Therefore, higher weight can be allocated to keywords appearing in the title, the beginning of the text and the end of the text, when the keyword weight is calculated, the importance degree of the part of speech is considered, and noun > verb > adjective and adverbs > other words are allocated when the weight is allocated. The frequency of the keywords is multiplied by the corresponding weight to obtain the score of each initial keyword, the scores are ranked from big to small, and a first preset number (for example, 10) of initial keywords are selected as target keywords.

In one embodiment, before the obtaining the text of the specific type from the preset database, the method further includes:

The method for extracting structured data such as news headlines, time and texts in real time can be realized by OCR recognition technology, the method for extracting structured data is not limited, the extracted structured data is stored in a preset database, and a user can conveniently obtain a specific type of text in a preset time period.

Step S20: and inputting the target keyword into a first model, judging whether the output result of the first model is a first result, if so, inputting the target keyword into a second model, judging whether the output result of the second model is a second result, and if so, acquiring parameter information of a plurality of initial objects corresponding to the second result.

In this embodiment, taking early warning in the insurance recommendation field as an example to describe the embodiment, inputting the target keyword into the first model, where the first model may be a prediction model that statistically analyzes a large amount of data of risk occurrence caused by disasters for a certain period of time, classifies out a corresponding disaster type or disaster influence size, determines whether an output result of the first model is a first result, and may be a type that the target keyword does not hit a corresponding disaster, or predicts that the target keyword does not have a corresponding disaster influence, and when the output result of the first model is the first result, inputs the target keyword into the second model, determines whether the output result of the second model is a second result, where the second model may be a model that performs clustering operation on market conditions (for example, market occupancy, product acceleration curve, price per label, label single-generation efficiency, etc.) of various policy products in the year by using a clustering algorithm, where the second result may be a policy product classification after the target keyword hits the second model, and the policy product classification may include: if the output result of the second model is the second result, acquiring parameter information of a plurality of initial objects (policy products) corresponding to the second result, wherein the parameter information comprises: price, type, applicable age interval, amount, etc. of policy products.

In one embodiment, the second model is obtained by performing a clustering operation on a preset sample object set based on a clustering algorithm, and the specific clustering operation includes:

The DBSCAN algorithm is a density-based clustering algorithm that generally assumes that the class can be determined by how tightly the sample is distributed. Samples of the same class are closely connected, that is, samples of the same class must exist in the vicinity of any sample of the class, and a clustering class is obtained by classifying the closely connected samples into one class, and a final clustering class result is obtained by classifying all groups of closely connected samples into different classes.

Further, as the selection of global parameters Eps and MinPts of the DBSCAN algorithm depends on manual intervention, after the data with uniform density distribution are arranged according to the ascending order of the k-dist curve, the point where the change amplitude of the curve begins to rise is manually selected as the Eps parameter, and the MinPts parameter is determined to be a fixed constant 4, the implementation process is complicated, and the implementation process depends on the manual intervention.

Therefore, reasonable global parameters Eps and MinPts can be determined in an adaptive mode, and a mode that a part of representative objects are selected to serve as seed objects to conduct class expansion in the area query process, and neighborhood objects of all core objects are not used as seeds to conduct class expansion is adopted. The flow is as follows:

(1) Adaptively determining global parameters Eps and MinPts;

(2) Classifying all sample points, respectively marking the sample points as core samples, boundary sample points and noise sample points, and deleting the marked noise sample points;

(3) Connecting all core points with the distance within the Eps distance, and classifying the core points into the same cluster;

(4) The core points in each cluster correspond to the selection of the seed representative object;

(5) Traversing the data set of the various policy product labels, carrying out regional query according to the selected representative object, and dividing the boundary points into clusters corresponding to the core points. If all points in the data set are processed, the algorithm ends.

Because the density measurement index is single, the data set is mainly aimed at data with insignificant cluster density difference. Calculating a distance distribution matrix DIST according to the input policy data set D _{n nx} The formula includes:

where n is the number of objects of the policy data set D. DIST (DIST) _{n nx} Is a real symmetric matrix of n rows and n columns, where each element represents the distance between object i and object j (i.e., the product policy) in data set D. Calculating DIST _{n nx} The values of each element in (c) are then arranged in ascending order row by row. With DIST _{n ix} Representing DIST _{n nx} For DIST of the value of column i in (3) _{n ix} Each row is arranged in ascending order to obtain KNN distribution;

performing curve fitting on the KNN distribution data obtained by ascending arrangement through a polynomial curve fitting formula, wherein the polynomial curve fitting formula is as follows:

wherein a, b, c, d, e is 5 adjacent sample points, x is the distance between the sample point and the abcde sample point, and the x after the second derivation is solved to obtain:

since the smaller value is a point close to the boundary, taking the larger value in the xsolution, discarding the smaller value and bringing it into the above polynomial can result in eps=f (x). Determining the value of MinPts, namely sequentially calculating the object number of the Eps-neighborhood of each point according to the statistical distribution characteristic of the data points in the field of each point, and then calculating the mathematical expectation of the data object, wherein the formula is as follows:

Where n is the number of objects, P, of the policy dataset D _i Points representing the Eps neighborhood at point i.

In one embodiment, the method further comprises:

when the output result of the first model is judged not to be the first result, sending first prompt information to a preset user side; and when the output result of the second model is judged not to be the second result, sending second prompt information to a preset user side.

When the output result of the first model is judged not to be the first result, the target keyword is indicated to hit the corresponding disaster type, or the target keyword is predicted to have the corresponding disaster influence, so that a corresponding early warning prompt can be sent to a manager, when the output result of the second model is judged not to be the second result, the target keyword is indicated not to be in the cluster of the existing policy products, and because the potential to plan the premium target or have a new positive influence on customer service, prompt information can be sent for a product decision maker to refer to.

Step S30: and screening a second preset number of target objects from the plurality of initial objects based on the parameter information and the preset dimensions of the plurality of initial objects, and pushing the target objects to a preset user group.

In this embodiment, a preset number of target objects may be screened from the multiple initial objects according to the duration dimension, the cost dimension, the acceleration dimension and the parameter information of the multiple initial objects (policy products), and the target objects are pushed to a preset user group, for example, by calculating and summarizing the historical policy, it is determined that the "vehicle bubble water insurance" policy class is 3 sets of policy products with the highest subsequent policy due to the client policy under the conditions of different insurance price intervals, payment intervals, insurance duration, insurance range, age range of the applicant, vehicle attribution and the like;

and calculating 3 sets of policy products with the optimal difference value in a certain requirement range by using a cost dimension to calculate the policy under the condition ranges of different ticket cost rates, price intervals and the like, and pushing the calculated policy products to a preset user group by using 3 sets of policy products with the fastest policy speed increasing proportion under different conditions by using a speed increasing dimension.

In one embodiment, the pushing modes include a first pushing mode and a second pushing mode, where the first pushing mode includes: when a request for acquiring an object is received from a user in a preset user group, carrying out identity verification on the user, and pushing the target object to the user when the identity verification passes;

The first pushing mode (for example, the user actively obtains) performs identity verification on the user when receiving a request for object obtaining sent by the user. For example, the device identifier included in the request is acquired, and whether the device identifier is a pre-bound device identifier (white list) is determined. And a second pushing mode (for example, the user passively receives), acquires the IP address of the user, and pushes the target object to the corresponding user.

Referring to fig. 2, a functional block diagram of an object pushing apparatus 100 based on multi-source data according to the present invention is shown.

The object pushing device 100 based on multi-source data can be installed in an electronic device. The multi-source data based object pushing device 100 may include a preprocessing module 110, a judging module 120, and a pushing module 130 according to the implemented functions. The module of the present invention may also be referred to as a unit, meaning a series of computer program segments capable of being executed by the processor of the electronic device and of performing fixed functions, stored in the memory of the electronic device.

In the present embodiment, the functions concerning the respective modules/units are as follows:

the preprocessing module 110 is configured to obtain a specific type of text from a preset database, perform preprocessing on the specific type of text to obtain a plurality of initial keywords, calculate a frequency of occurrence of each initial keyword in the specific type of text, and screen a first preset number of target keywords based on the frequency and a preset weight.

In one embodiment, before the obtaining the text of the specific type from the preset database, the preprocessing module is further configured to:

The judging module 120 is configured to input the target keyword into the first model, judge whether an output result of the first model is a first result, if yes, input the target keyword into the second model, judge whether an output result of the second model is a second result, and if yes, obtain parameter information of a plurality of initial objects corresponding to the second result.

(1) Adaptively determining global parameters Eps and MinPts;

In one embodiment, the judging module is further configured to:

The pushing module 130 is configured to screen a second preset number of target objects from the plurality of initial objects based on the parameter information and the preset dimensions of the plurality of initial objects, and push the target objects to a preset user group.

Referring to fig. 3, a schematic diagram of a preferred embodiment of an electronic device 1 according to the present invention is shown.

The electronic device 1 includes, but is not limited to: memory 11, processor 12, display 13, and network interface 14. The electronic device 1 is connected to a network through a network interface 14 to obtain the original data. The network may be a wireless or wired network such as an Intranet (Intranet), the Internet (Internet), a global system for mobile communications (Global System of Mobile communication, GSM), wideband code division multiple access (Wideband Code Division Multiple Access, WCDMA), a 4G network, a 5G network, bluetooth (Bluetooth), wi-Fi, or a call network.

The memory 11 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 11 may be an internal storage unit of the electronic device 1, such as a hard disk or a memory of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are equipped in the electronic device 1. Of course, the memory 11 may also comprise both an internal memory unit of the electronic device 1 and an external memory device. In this embodiment, the memory 11 is generally used to store an operating system and various application software installed in the electronic device 1, such as program codes of the object pushing program 10 based on multi-source data. Further, the memory 11 may be used to temporarily store various types of data that have been output or are to be output.

Processor 12 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 12 is typically used for controlling the overall operation of the electronic device 1, e.g. performing data interaction or communication related control and processing, etc. In this embodiment, the processor 12 is configured to execute the program code stored in the memory 11 or process data, for example, execute the program code of the object pushing program 10 based on multi-source data.

The display 13 may be referred to as a display screen or a display unit. The display 13 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an Organic Light-Emitting Diode (OLED) touch device, or the like in some embodiments. The display 13 is used for displaying information processed in the electronic device 1 and for displaying a visual work interface, for example displaying the results of data statistics.

The network interface 14 may alternatively comprise a standard wired interface, a wireless interface, such as a WI-FI interface, which network interface 14 is typically used for establishing a communication connection between the electronic device 1 and other electronic devices.

Fig. 3 shows only the electronic device 1 with components 11-14 and the object pushing program 10 based on multi-source data, but it should be understood that not all shown components are required to be implemented, and that more or fewer components may alternatively be implemented.

Optionally, the electronic device 1 may further comprise a user interface, which may comprise a Display (Display), an input unit such as a Keyboard (Keyboard), and a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an Organic Light-Emitting Diode (OLED) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 1 and for displaying a visual user interface.

The electronic device 1 may further comprise Radio Frequency (RF) circuits, sensors and audio circuits etc., which are not described here.

In the above embodiment, the processor 12 may implement the following steps when executing the object pushing program 10 based on multi-source data stored in the memory 11:

The storage device may be the memory 11 of the electronic device 1, or may be another storage device communicatively connected to the electronic device 1.

For a detailed description of the above steps, please refer to the above-mentioned functional block diagram of fig. 2 regarding an embodiment of the object pushing apparatus 100 based on multi-source data and the description of fig. 1 regarding a flowchart of an embodiment of the object pushing method based on multi-source data.

Furthermore, the embodiment of the invention also provides a computer readable storage medium, which can be nonvolatile or volatile. The computer readable storage medium may be any one or any combination of several of a hard disk, a multimedia card, an SD card, a flash memory card, an SMC, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, etc. The computer readable storage medium includes a storage data area and a storage program area, the storage data area stores data created according to the use of the blockchain node, the storage program area stores an object pushing program 10 based on multi-source data, and the object pushing program 10 based on multi-source data realizes the following operations when being executed by a processor:

The embodiment of the computer readable storage medium of the present invention is substantially the same as the embodiment of the object pushing method based on multi-source data, and will not be described herein.

In another embodiment, in the object pushing method based on multi-source data provided by the present invention, in order to further ensure the privacy and security of all the data that appear, all the data may also be stored in a node of a blockchain. Such as target keywords and parameter information, etc., which may be stored in the blockchain node.

It should be noted that, the blockchain referred to in the present invention is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm, etc. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

It should be noted that, the foregoing reference numerals of the embodiments of the present invention are merely for describing the embodiments, and do not represent the advantages and disadvantages of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, an electronic device, or a network device, etc.) to perform the method according to the embodiments of the present invention.

The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.

Claims

1. An object pushing method based on multi-source data, which is characterized in that the method comprises the following steps:

screening a second preset number of target objects from the plurality of initial objects based on the parameter information and preset dimensions of the plurality of initial objects, and pushing the target objects to a preset user group;

Wherein the preprocessing the text of the specific type to obtain a plurality of initial keywords comprises: performing word segmentation operation on the text of the specific type to obtain a plurality of segmented words, and extracting a plurality of initial keywords from the segmented words based on a TF-IDF algorithm;

the screening the first preset number of target keywords based on the frequency number and the preset weight comprises the following steps: assigning associated weights to initial keywords of preset parts of speech in the plurality of initial keywords based on a preset assignment rule, calculating to obtain scores of the initial keywords based on the frequency and the weights, sequencing the scores from large to small, and selecting a first preset number of initial keywords as the target keywords;

the second model is obtained by performing clustering operation on a preset sample object set based on a clustering algorithm, and the specific clustering operation comprises: setting a density radius between sample objects and the number of sample objects with the smallest density radius, and based on the density radius and the number of sample objects with the smallest density, iteratively calculating to obtain a core sample object, a sample object with the reachable density and an edge sample object from all sample objects, obtaining the sample object with the reachable density of the core sample object, and updating a cluster corresponding to the core sample object by utilizing the sample object with the reachable density obtained by the iterative calculation;

The pushing modes comprise a first pushing mode and a second pushing mode, and the first pushing mode comprises: when a request for acquiring an object is received from a user in a preset user group, carrying out identity verification on the user, and pushing the target object to the user when the identity verification passes; the second pushing mode includes: and acquiring the related information of the preset user group, wherein the related information comprises the IP address of each user, and pushing the target object to the preset user group according to the related information.

2. The multi-source data based object pushing method of claim 1, wherein prior to the retrieving the specific type of text from the preset database, the method further comprises:

3. The multi-source data based object pushing method of claim 1, wherein the method further comprises:

4. A multi-source data based object pushing apparatus for implementing the multi-source data based object pushing method according to any one of claims 1 to 3, the apparatus comprising:

5. An electronic device, the electronic device comprising:

at least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein,

the memory stores a program executable by the at least one processor to enable the at least one processor to perform the multi-source data-based object pushing method of any one of claims 1 to 3.

6. A computer-readable storage medium, wherein the computer-readable storage medium stores a multi-source data-based object pushing program, which when executed by a processor, implements the multi-source data-based object pushing method according to any one of claims 1 to 3.