CN111695030A - Parameter acquisition method and device, electronic equipment and computer readable storage medium - Google Patents
Parameter acquisition method and device, electronic equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN111695030A CN111695030A CN202010340907.8A CN202010340907A CN111695030A CN 111695030 A CN111695030 A CN 111695030A CN 202010340907 A CN202010340907 A CN 202010340907A CN 111695030 A CN111695030 A CN 111695030A
- Authority
- CN
- China
- Prior art keywords
- preset
- user
- determined
- interest value
- wifi
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—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
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The invention provides a parameter acquisition method, a parameter acquisition device, electronic equipment and a computer-readable storage medium, and belongs to the technical field of electronic equipment. According to the method, for a user to be determined, a first interest value of the user to be determined for a preset content category can be obtained, wherein the first interest value is determined by using user characteristics of the user to be determined and a preset classification model, a second interest value of the user to be determined for the preset content category is obtained according to a place where the user to be determined appears in a preset time period and a preset place type corresponding to the preset content category, and a final interest value of the user to be determined for the preset content category is calculated according to the first interest value and the second interest value. Therefore, the method for determining the final interest value reflecting the user interest is combined with the method for determining the final interest value reflecting the user interest, and the accuracy of the determined user interest can be improved to a certain extent.
Description
Technical Field
The present invention relates to the field of electronic devices, and in particular, to a parameter obtaining method and apparatus, an electronic device, and a computer-readable storage medium.
Background
With the continuous development of computer technology, computers are more and more widely applied. In order to facilitate a user to quickly acquire interesting content through an electronic device, the user often needs to be recommended with the interesting content according to the personal interest of the user.
In the prior art, when determining the user interest, the user interest level of a preset content category is often calculated according to fixed information preset by the user, and direct recommendation is performed based on the user interest level of the preset content category. Therefore, the method for determining the user interest directly based on the fixed information is low in accuracy, and further poor in recommendation effect.
Disclosure of Invention
The invention provides a parameter acquisition method, a parameter acquisition device, electronic equipment and a computer-readable storage medium, which are used for solving the problem of low accuracy of determined user interest.
In a first aspect of the present invention, a parameter obtaining method is provided, where the method includes:
for a user to be determined, acquiring a first interest value of the user to be determined in a preset content category; the first interest value is determined by utilizing the user characteristics of the user to be determined and a preset classification model;
acquiring a second interest value of the user to be determined on the preset content category according to the place where the user to be determined appears in the preset time period and the preset place type corresponding to the preset content category;
and calculating the final interest value of the user to be determined in the preset content category according to the first interest value and the second interest value.
In a second aspect of the present invention, there is also provided a parameter obtaining apparatus, including:
the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a first interest value of a user to be determined on a preset content category; the first interest value is determined by utilizing the user characteristics of the user to be determined and a preset classification model;
the second obtaining module is used for obtaining a second interest value of the user to be determined on the preset content category according to the place where the user to be determined appears in the preset time period and the preset place type corresponding to the preset content category;
and the calculating module is used for calculating the final interest value of the user to be determined in the preset content category according to the first interest value and the second interest value.
In yet another aspect of the present invention, there is also provided a computer-readable storage medium having stored therein instructions, which when run on a computer, cause the computer to execute any one of the above-described parameter acquisition methods.
In yet another aspect of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform any of the above-described parameter acquisition methods.
The parameter obtaining method provided by the embodiment of the invention obtains a first interest value of a user to be determined for a preset content category by obtaining the first interest value of the user to be determined, wherein the first interest value is determined by using a user characteristic of the user to be determined and a preset classification model, the user characteristic is determined according to user personal information of the user to be determined, then, a second interest value of the user to be determined for the preset content category can be obtained according to a place where the user to be determined appears in a preset time period and a preset place type corresponding to the preset content category, and finally, a final interest value of the user to be determined for the preset content category is calculated according to the first interest value and the second interest value. The method and the device have the advantages that the position where the user to be determined appears in the preset time length can reflect the nearest short-term interest of the user to be determined to a certain extent, so that the method and the device can determine the final interest value reflecting the interest of the user by using the first interest value determined according to the personal information of the user and combining the second interest value determined according to the occurrence parameter, the accuracy of the determined final interest value reflecting the interest of the user can be improved to a certain extent, and therefore when the user is recommended based on the final interest value in the following process, the recommended content to the user can meet the requirements of the user better, and the recommendation effect is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
Fig. 1 is a flowchart illustrating steps of a parameter obtaining method according to an embodiment of the present invention;
FIG. 2-1 is a flow chart illustrating steps of another parameter obtaining method according to an embodiment of the present invention;
fig. 2-2 is a schematic diagram of a parameter obtaining method according to an embodiment of the present invention;
FIG. 3-1 is a block diagram of a parameter obtaining apparatus according to an embodiment of the present invention;
fig. 3-2 is a block diagram of another parameter obtaining apparatus provided in the embodiment of the present invention;
fig. 4 is a block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention.
Fig. 1 is a flowchart of steps of a parameter obtaining method according to an embodiment of the present invention, and as shown in fig. 1, the method may include:
In the embodiment of the invention, the user to be determined can be a user needing to determine the user interest, and the user to be determined can be any user in the network. For example, the user to be determined may be a user in a video playback application. The user characteristics of the user to be determined are determined according to the user personal information of the user to be determined. The user personal information may be information related to the user to be determined, for example, a physical health condition of the user to be determined, an annual income of the user to be determined, the number of family members of the user to be determined, an age of the family members of the user to be determined, and the like. The personal information of the user can be crawled from a network by the electronic equipment, and can also be collected to the user to be determined through a displayed information collection page. Further, the user personal information of each dimension can be vectorized, and then the vector obtained after vectorization is determined as the user characteristic of the user to be determined. The preset content category may be a category of content to be subsequently recommended, wherein the content to be recommended may be an article, a video, music, or the like.
Further, the preset classification model may be obtained by training the initial classification model in advance through sample data based on a big data machine learning method. The preset classification model can predict the interest degree of the user to be determined in each preset content category according to the user characteristics, and finally, outputs the probability value of the interest of the user to be determined in each preset content category. The greater the probability value is, the greater the interest degree of the user to be determined in the preset content category may be.
The sample data may include a plurality of sample pairs consisting of sample user characteristics and true class values of the sample users. And the real category value is used for representing the probability value of the interest degree of the sample user in the preset content category. The sample users may be people who are interested in the preset content category screened from the network according to the characteristics that the users interested in the preset content category may have. For example, assuming that the preset content category is a mother-infant product, users whose family members include children of 0-3 years old can be screened from the network as sample users.
In the screening, a material filling page is displayed, then the content input by the user from the page is received, and then the screening is carried out according to the content. When the sample user characteristics of the sample user are obtained, the personal information of the sample user, such as basic personal information of the sample user, viewing behavior information, member service information, information reported by application software installed in used electronic equipment, and related information for searching, may be obtained according to the personal information of the sample user crawled from the network. Then, vectorizing the information to obtain the sample user characteristics. It should be noted that, in the embodiment of the present invention, the sample user characteristics may also be supplemented into a preset user image database to provide more data for depicting the user image, so as to improve the depicting degree of the user image and increase the depicting effect of the user image.
102, acquiring a second interest value of the user to be determined in the preset content category according to the place where the user to be determined appears in the preset time period and the preset place type corresponding to the preset content category.
The preset time period in the embodiment of the present invention may be set according to an actual requirement, for example, the preset time period may be a time period formed by the preset time period, where the preset time period may be set according to a requirement, for example, the preset time period may be 3 days, 5 days, or 7 days, and so on. The preset location type corresponding to the preset content category may be preset according to an actual situation, and the preset location type corresponding to the preset content category may be a type of a location that a user may go to when the user is interested in the preset content category. One or more preset location types corresponding to one preset content category may be provided.
Since the first interest value is determined based on the user personal information, the first interest value may be considered to reflect the user interest of the user to be determined from the user personal information level. The place where the user goes often changes due to the influence of recent interests, for example, the user suddenly likes to exercise recently, and the user often goes to the place where the gymnasium capable of exercising is located recently, so that the appearance of the user to be determined in different places can reflect the recent interests of the user to a certain extent. Based on this, in order to improve the accuracy of the determined user interest, in this step, a second interest value of the user to be determined for the preset content category may be further determined according to a place where the user to be determined appears within the preset time period and a preset place type corresponding to the preset content category.
In the embodiment of the invention, the first interest value can reflect the user interest of the user to be determined from the user personal information level, and the frequency of the change of the user personal information is low, namely, the first interest value can reflect the interest degree of the user to be determined in the preset content category under the relatively stable fixed interest, and the second interest value can reflect the interest degree of the user to be determined in the preset content category under the recent interest, so that the first interest value and the second interest value can be combined to determine the final interest value. The final interest value is combined with the first interest value and the second interest value, so that the user interest of the user to be determined can be reflected more comprehensively to a certain extent, and the accuracy of the determined user interest can be improved.
In summary, the parameter obtaining method provided in the embodiment of the present invention may obtain, for a user to be determined, a first interest value of the user to be determined in a preset content category, where the first interest value is determined by using a user characteristic of the user to be determined and a preset classification model, the user characteristic is determined according to user personal information of the user to be determined, then, a second interest value of the user to be determined in the preset content category may be obtained according to a place where the user to be determined appears in a preset time period and a preset place type corresponding to the preset content category, and finally, a final interest value of the user to be determined in the preset content category is calculated according to the first interest value and the second interest value. The method and the device have the advantages that the position where the user to be determined appears in the preset time length can reflect the nearest short-term interest of the user to be determined to a certain extent, so that the method and the device can determine the final interest value reflecting the interest of the user by using the first interest value determined according to the personal information of the user and combining the second interest value determined according to the occurrence parameter, the accuracy of the determined final interest value reflecting the interest of the user can be improved to a certain extent, and therefore when the user is recommended based on the final interest value in the following process, the recommended content to the user can meet the requirements of the user better, and the recommendation effect is improved.
Fig. 2-1 is a flowchart of steps of another parameter obtaining method provided in an embodiment of the present invention, and as shown in fig. 2-1, the method may include:
In this step, when the first interest value is obtained, the user characteristics may be used as input of a preset classification model, and the preset classification model is used to generate the first interest value; the user characteristics are determined based on personal information of the user to be determined. Specifically, the probability value that the user to be determined is interested in the preset content category may be determined according to the user characteristics by using the preset classification model. And then determining the probability value as a first interest value, or determining a numerical value matched with the probability value as the first interest value according to a preset corresponding relation between the probability value and the numerical value. Therefore, the first interest value is generated in real time through the preset classification model, and the accuracy of the first interest value can be ensured to a certain extent.
Alternatively, the obtaining of the first interest value may be implemented by: according to the user Identification (ID) of the user to be determined, searching a first interest value corresponding to the user ID from a preset first interest value database to obtain a first interest value of the user to be determined on a preset content category; the preset first interest value database stores first interest values of different users for preset content categories, and the first interest values stored in the first interest value database are generated in advance according to preset classification models and user characteristics of the users. Therefore, the first interest value can be obtained by directly searching, and the obtaining efficiency of the first interest value can be improved to a certain extent.
Specifically, after the preset classification model is obtained through training, the preset classification model is used to determine first interest values corresponding to different users according to user characteristics corresponding to the different users in the network, and the first interest values are stored in the preset first interest value database. That is, global prediction is performed, and accordingly, in this step, the first interest value corresponding to the user to be determined may be directly searched from the predetermined first interest values. Because the processing resources consumed by the model training and the global prediction are more, in the embodiment of the present invention, the two operations can be implemented by other devices, so as to save the processing resources of the device executing the parameter obtaining method provided by the embodiment of the present invention, and further ensure the stable operation of the device. It should be noted that, in order to ensure the accuracy of the first interest value stored in the preset first interest value database, in the embodiment of the present invention, the user information of different users in the network may be re-acquired according to a preset period, then user characteristics of different users are generated based on the user information of different users, then, the preset classification model is used to re-determine the first interest value corresponding to different users according to the user characteristics of the different users, and the first interest value stored in the preset first interest value database is updated based on the re-determined first interest value.
Specifically, this step can be realized by the following substeps (1) to (3):
substep (1): and acquiring the WIFI identifier of the historical wireless local area network WIFI network connected by the user to be determined in the preset time period.
In an actual application scenario, a user often has a large dependence on a network, and each time the user arrives at a location, the user often connects to a Wireless-Fidelity (WIFI) of the location. That is, it may be considered that the user is present at the location when the user is connected to the WIFI network. Therefore, in the embodiment of the invention, the WIFI identifier of the WIFI network connected by the user to be determined within the preset time length can be obtained, so as to determine the place where the user to be determined appears.
Specifically, the WIFI identifier of the WIFI network may be an identifier used for representing the WIFI network, and when the WIFI identifier of the connected historical WIFI network is obtained, a scanned WIFI list when the user starts the WIFI connection function within a preset time period may be obtained. And then analyzing each WIFI network displayed in the WIFI list to acquire a WIFI identifier of the WIFI network. And then, extracting a WIFI identifier corresponding to the WIFI network to be determined to which the user selects to connect. It should be noted that, in order to increase the number of the obtained WIFI identifiers and provide more determination bases for subsequent operations, in the embodiment of the present invention, the WIFI identifiers of the WIFI networks whose distances from the WIFI networks connected to the user to be determined are smaller than the preset distance threshold may be extracted from the WIFI identifiers of the WIFI networks corresponding to the WIFI list by extending through a heuristic rule based on spatial invariance of the WIFI and the characteristic that the locations in the same area may belong to the same location type. Therefore, the data are expanded by extracting the WIFI identification of the WIFI network with the short distance, more data bases can be provided for subsequent operation, and the accuracy of determining the operation is improved to a certain extent.
Substep (2): and dividing the historical WIFI network into WIFI networks which are matched with the historical WIFI network and are included by preset place types according to the corresponding relation among the WIFI identifications, the preset WIFI identifications and the preset place types.
In this step, the preset location type may be set according to actual requirements, and for example, the preset location type may be a school, a hospital, an office area, a natural scenic spot, a business place, a sport place, and the like. The WIFI identifier may include a WIFI name and a WIFI address. Due to the fact that the WIFI names and the WIFI addresses are not accurate enough due to error influence, the WIFI identifiers comprise the WIFI names and the WIFI addresses at the same time, and matching precision is higher when matching is conducted on the basis of the WIFI identifiers in subsequent steps. Of course, the WIFI identifier may also be set to include one of them, which is not limited in the embodiment of the present invention.
Further, the corresponding relationship between the preset WIFI identifier and the preset location type may be generated in advance through the following steps a to D:
and step A, for the preset WIFI network, segmenting the WIFI name of the preset WIFI network to obtain name words.
In an actual application scenario, the WIFI names of the WIFI networks often include information that can reflect the location where the WIFI network is located, for example, the WIFI names of the WIFI networks of schools often include names of the schools, and the WIFI names of the WIFI networks of malls often include names of the malls. Therefore, when the corresponding relation is generated, the name words can be obtained by using the WIFI names of the preset WIFI network, and the preset place types corresponding to the preset WIFI network can be accurately determined according to the name words in the subsequent steps. Meanwhile, the WIFI name is always public, so that the confidentiality is low. Therefore, in the embodiment of the invention, the preset place type corresponding to the preset WIFI network is determined on the basis of the WIFI name, so that the determination cost can be reduced to a certain extent.
Wherein, the name word can be a word contained in the WIFI name. When performing word segmentation, word segmentation can be performed based on the NLP technology. Specifically, word-by-word traversal can be performed through a common word segmentation library, for example, a common dictionary, and the like, all words in the common word segmentation library are respectively traversed and matched in the WIFI name according to the arrangement sequence, if matching is successful, the current word is determined to be a word included in the WIFI name, and the process is repeated until all words in the common word segmentation library are matched once, and all words included in the WIFI name are determined, and the words are name words. It should be noted that the name word may include characters such as chinese, english, and numbers.
And step B, matching the name words with preset word packets corresponding to each preset place type to determine a first number of words matched with the name words in the preset word packets, and matching the WIFI addresses of the preset WIFI network with preset addresses corresponding to each preset place type to determine a second number of addresses matched with the name words in the preset addresses.
In this step, the preset word packet and the preset address corresponding to the preset location type may be predetermined. Specifically, the preset word packet and the preset address can be obtained through the following operations: and aiming at any one preset place type, acquiring a WIFI network of which the set place belongs to the preset place type as a sample WIFI network. Specifically, the set location of each WIFI network can be determined from WIFI data reported through the user behavior log, and then whether the WIFI network belongs to the preset location type or not is determined according to the set location of the WIFI network. If so, the WIFI network may be determined to be the sample WIFI network.
It should be noted that certain dirty data may exist in the WIFI data reported by the user behavior log, for example, the dirty data includes a WIFI network with an illegal address, an abnormal WIFI network connected by too many devices, and the like. In order to improve the quality of the screened sample WIFI network, in the embodiment of the present invention, the WIFI data may be filtered first, so as to delete dirty data therein. Like this, through deleting dirty data, can avoid dirty data to be selected as sample WIFI network, and then can improve the selection quality to a certain extent. When filtering is carried out specifically, the address of the sample WIFI can be compared with the address of a preset illegal WIFI network, and if the address of the sample WIFI is matched with the address of the illegal WIFI network, the sample WIFI is deleted. Or determining the number of the sample WIFI connection devices, and deleting the sample WIFI if the number is larger than a preset number threshold. Or comparing the address of the sample WIFI with the preset address of the illegal WIFI network, determining the number of the sample WIFI connection devices, and deleting the sample WIFI if the address of the sample WIFI is matched with the address of the illegal WIFI network and the number is larger than a preset number threshold value. Then, a WIFI address of the sample WIFI network may be obtained to serve as a preset address corresponding to the preset location type. Specifically, the WIFI address may be a Media Access Control (MAC) address of the WIFI network. When the address is obtained, the MAC address of the sample WIFI network can be searched from the WIFI data, and the searched MAC address is used as a preset address corresponding to the preset place type. By analogy, when a plurality of sample WIFI networks exist, a plurality of preset addresses can be obtained. Further, word segmentation processing can be performed on the WIFI names of the sample WIFI network to obtain sample words; and selecting preset words from the sample words according to the occurrence times of the sample words so as to generate the preset word packet. Specifically, the sample word may be a word included in the WIFI name of the sample WIFI network, and the specific implementation manner of word segmentation may refer to the related description in the foregoing content, which is not repeated herein in the embodiments of the present invention. When the preset words are selected from the sample words according to the occurrence times of the sample words to generate the preset word package, the occurrence times of each sample word may be counted first, and then the TGI preference of the sample words relative to the preset location type may be calculated according to the occurrence times and the total number of the included sample words and by the following formula.
TGI preference ═ standard number [ number of occurrences/total number of sample words ].
The standard number may be 100.
Next, the sample word whose TGI preference is at the head, that is, the sample word whose TGI preference is greater than the preset threshold, is determined as the preset word. And finally, forming a word packet by using the selected preset words to obtain the preset word packet. Because the TGI preference can more accurately reflect the possibility that the sample word appears in the name of the WIFI network set in the location of the preset location type, for example, "EDU" often appears in the preset location type: in the WIFI names of "school", the "hospital" often appears in the preset location types: in the WIFI name of "hospital". Then, the TGI preference corresponding to these words tends to be large. Therefore, the accuracy of the generated preset word packet can be improved to a certain extent by firstly calculating the TGI preference and selecting according to the TGI preference. Meanwhile, the preset word is used for representing the type of the preset place, and the subsequent steps are based on the mode of matching the preset word, so that fuzzy matching can be realized, the limitation of matching can be reduced to a certain extent, and the matching effect is improved. Of course, the sample words with the occurrence times larger than the preset threshold value can also be directly selected as the preset words, so that the generation of the word packet can be realized without additional calculation, and the processing resources can be saved to a certain extent.
Further, when the first number is determined, the name words and the words in the preset word package corresponding to the preset location type may be compared one by one, and then the number of the same words is counted to obtain the first number. When the second number is determined, the WIFI addresses of the preset WIFI network and the preset addresses corresponding to the preset place types can be compared one by one, and then the number of the same addresses is counted to obtain the second number. It should be noted that the address of the WIFI network is often unique, and therefore, the second number is often 0 or 1. For example, when the preset WIFI network belongs to a network set in a location corresponding to the preset location type, the second number may be 1, and when the preset WIFI network does not belong to a network set in a location corresponding to the preset location type, the second number may be 0. Further, because the determined address may have an error, in the embodiment of the present invention, it may be determined that the two addresses are the same when the similarity of the addresses is greater than the preset similarity threshold, so as to improve the fault tolerance of the scheme.
And step C, determining the corresponding preset place types of which the first quantity is not less than a first preset threshold value and the second quantity is not less than a second preset threshold value as the preset place types corresponding to the preset WIFI network.
In this step, the first preset threshold and the second preset threshold may be set according to actual conditions. The first preset threshold and the second preset threshold may be the same or different. Further, if the corresponding first number is not less than the first preset threshold and the second number is not less than the second preset threshold, the location set by the preset WIFI network may be considered as belonging to the preset location type, and therefore, it may be determined that the preset location type is determined as the preset location type corresponding to the preset WIFI network. For example, assuming that the first preset threshold is 2, the second preset threshold is 1, the first number of the preset WIFI networks corresponding to the preset location type "school" is 2, and the second number is 1, it may be determined that the preset location type to which the location set by the preset WIFI network belongs is school.
And D, establishing a corresponding relation between the preset WIFI identification and the preset place type according to the WIFI identification of the preset WIFI network and the preset place type corresponding to the preset WIFI network.
In this step, the WIFI identifier of the preset WIFI network may be associated with the preset location type corresponding to the preset WIFI network, so as to obtain the corresponding relationship.
Further, when the historical WIFI network is divided into the WIFI networks included in the preset location types, the WIFI identifiers of the historical WIFI networks can be matched with the WIFI identifiers in the corresponding relations, and if the historical WIFI networks are matched with the WIFI identifiers, for example, the historical WIFI networks can be divided into the WIFI networks included in the preset location types. Through division, different preset place types including different historical WIFI networks can be obtained.
Substep (3): and counting the sum of the connection times and/or the connection duration of the user to be determined and the WIFI network included in the target preset place type to serve as the occurrence parameter.
In an actual application scene, when a user appears at a place, the user can be connected with the WIFI network sequentially arranged at the place, and the duration of the place is the connection duration of the WIFI network. Therefore, in this step, the sum of the connection times and/or the connection duration of the user to be determined and the WIFI network included in the target preset location type may be obtained, and specifically, the sum of the time information and/or the sum of the duration information may be extracted from the historical connection information of the user to be determined to serve as the occurrence parameter. The sum of the connection times and the sum of the connection durations may represent the total number of times of occurrence and the total duration of occurrence of the user to be determined at the location of the preset location type.
Because the difficulty in acquiring the WIFI connection information of the user is low, in the embodiment of the invention, the occurrence parameters of the place where the user to be determined appears are indirectly determined based on the WIFI network connected with the user to be determined, so that the parameter determination cost can be reduced to a certain extent, and the determination difficulty can be reflected. Of course, in the embodiment of the present invention, information of a Global Positioning System (GPS) used by the user to be determined within a preset time period may also be directly obtained, and then, a place where the user to be determined appears may be determined according to analysis of the GPS information. And classifying different places in advance to obtain a corresponding relation between a preset place and a preset place type, and finally determining occurrence parameters by combining the corresponding relation between the preset place type and the place where the user to be determined appears.
In this step, the preset parameter threshold may be set according to actual conditions. Under the condition that the user is in different interest degrees with respect to the preset content categories, the situations of the target preset location type occur differently, so that the second interest value can be determined by combining the occurrence parameter and the preset parameter threshold value, so that the recent interest of the user to be determined is reflected by the second interest value.
Specifically, the preset parameter may be determined according to the occurrence of the user who is interested in the content of the preset content category corresponding to the target preset location type at the location of the target preset location type. For example, behavior characteristics of users who are really interested in the content of the preset content category may be analyzed, and the number of times and duration that the users will connect to the WIFI network set in the place belonging to the preset place type corresponding to the preset content category within the preset duration may be determined, for example, the behavior characteristics of the users may be determined for the preset content category: the users interested in the 'business articles' connect the places of the preset place types corresponding to the business articles, such as the times and the duration of the WIFI network set in the airport. And then calculating the number average and the duration average, and finally determining the number average and the duration average as preset parameter thresholds.
Further, the absolute value of the difference between the occurrence parameter and the preset parameter threshold may be calculated first when determining the second interest value. Specifically, the difference between the occurrence parameter and the preset parameter threshold may be calculated first, and then the absolute value of the difference may be calculated to obtain the absolute value of the difference between the occurrence parameter and the preset parameter threshold. When the difference between the occurrence parameter and the preset parameter threshold is calculated, the occurrence parameter may be used as the subtree, and the preset parameter threshold may be used as the subtree.
Then, the second interest value is determined based on the absolute value. Specifically, when the second interest value is determined, the determination may be performed in a manner that the second interest value is negatively correlated with the absolute value when the occurrence parameter is not greater than the preset parameter threshold, that is, a smaller second interest value is set for a user to be determined whose absolute value is larger. And under the condition that the occurrence parameter is larger than the preset parameter threshold, determining according to a positive correlation mode of the second interest value and the absolute value, namely, setting a larger second interest value for the user to be determined with a larger absolute value. In this way, it can be ensured that the determined second interest value can accurately reflect the interest of the user to be determined.
And 204, calculating a final interest value of the user to be determined in the preset content category according to the first interest value and the second interest value.
In this step, the final interest value may be determined according to the first interest value, the second interest value, and the preset interest value weight. The preset interest value weight may include a preset interest value weight corresponding to the first interest value and a preset interest value weight corresponding to the second interest value. When determining the final interest value, a preset interest value weight product of the first interest value and a preset interest value weight product of the second interest value and the second interest value may be calculated, and finally, a sum of the two products may be determined as the final interest value. Wherein, the preset interest value weight can be preset by the following method: firstly, setting a default interest value weight, then respectively determining the accuracy of the first interest value and the second interest value through a plurality of experiments, and finally, increasing the preset interest value weight corresponding to the first interest value and reducing the preset interest value weight corresponding to the second interest value until the proportion of the preset interest value weight is matched with the accuracy of the interest value.
Specifically, in this step, the content in the preset content category with the highest final interest value may be recommended to the user to be determined. Assuming that the preset content category with the highest final interest value is "mother-infant", the information of the mother-infant product may be used as recommendation information, and the information of the mother-infant product may be output to the user to be determined, for example, the information of the mother-infant product is displayed on a screen of a device used by the user to be determined. Of course, other recommendations may be made, and the embodiment of the present invention is not limited thereto.
For example, fig. 2-2 is a schematic diagram of a parameter obtaining method according to an embodiment of the present invention, as shown in fig. 2-2, mode discovery may be performed on WIFI data, that is, a preset word packet and a preset address corresponding to a preset location type are determined, and then, after the mode discovery, a corresponding relationship between a preset WIFI identifier and the preset location type is established according to a discovery result "commercial WIFI category", that is, the preset word packet and the preset address corresponding to the preset location type, where the preset location type may include a residence, a school, a company, an airport, a hospital, and the like. A second interest value is then determined based on the correspondence. And performing model prediction based on the personal information of the user to determine a first interest value. Wherein the first interest value may be determined based on information such as age, gender, life stage, marital status, behavioral patterns, etc. The two are then pattern-combined to determine a final interest value for the preset content category, i.e. a final interest value is calculated based on both. For example, final interest values for the categories of mother and infant, car, game, finance, health, luxury, etc. may be calculated.
In summary, the parameter obtaining method provided in the embodiment of the present invention may obtain, for a user to be determined, a first interest value of the user to be determined in a preset content category, where the first interest value is determined by using a user characteristic of the user to be determined and a preset classification model, the user characteristic is determined according to user personal information of the user to be determined, then, according to a place where the user to be determined appears in a preset time period, an occurrence parameter of the user to be determined in a place of a preset place type is obtained, according to the occurrence parameter and a preset parameter threshold corresponding to a target preset place type, a second interest value of the user to be determined in the preset content category is calculated, and finally, according to the first interest value and the second interest value, a final interest value of the user to be determined in the preset content category is calculated. The method comprises the steps of determining a final interest value representing the user interest by utilizing the first interest value determined according to the personal information of the user and combining the second interest value determined according to the occurrence parameter, and improving the accuracy degree of the determined final interest value capable of accurately representing the user interest.
Fig. 3-1 is a block diagram of a parameter obtaining apparatus according to an embodiment of the present invention, and as shown in fig. 3-1, the apparatus 30 may include:
a first obtaining module 301, configured to obtain, for a user to be determined, a first interest value of the user to be determined in a preset content category; the first interest value is determined by using the user characteristics of the user to be determined and a preset classification model.
A second obtaining module 302, configured to obtain a second interest value of the user to be determined in the preset content category according to a place where the user to be determined has appeared in a preset time period and a preset place type corresponding to the preset content category.
A calculating module 303, configured to calculate a final interest value of the to-be-determined user in the preset content category according to the first interest value and the second interest value.
Optionally, fig. 3-2 is a block diagram of another parameter obtaining apparatus provided in the embodiment of the present invention, and as shown in fig. 3-2, the apparatus 30 further includes:
an output module 304, configured to obtain recommendation information of the user to be determined according to the final interest value of the user to be determined in the preset content category, and output the recommendation information to the user to be determined.
Optionally, the second obtaining module 302 is specifically configured to:
acquiring occurrence parameters of the to-be-determined user in a place of a target preset place type according to the place where the to-be-determined user appears in a preset time period; the target preset place type is a preset place type corresponding to the preset content type.
And calculating a second interest value of the user to be determined in the preset content category according to the occurrence parameter and a preset parameter threshold corresponding to the type of the target preset place.
Optionally, the second obtaining module 302 is further specifically configured to:
and acquiring the WIFI identifier of the historical wireless local area network WIFI network connected by the user to be determined in the preset time period.
And dividing the historical WIFI network into WIFI networks which are matched with the historical WIFI network and are included by preset place types according to the corresponding relation among the WIFI identifications, the preset WIFI identifications and the preset place types.
And counting the sum of the connection times and/or the connection duration of the user to be determined and the WIFI network included in the target preset place type to serve as the occurrence parameter.
Optionally, the WIFI identifier includes a WIFI name and a WIFI address; the second obtaining module 302 is further specifically configured to:
and for the preset WIFI network, segmenting the WIFI name of the preset WIFI network to obtain name words.
Matching the name words with preset word packets corresponding to each preset place type to determine a first number of words in the preset word packets, which are matched with the name words, and matching the WIFI address of the preset WIFI network with a preset address corresponding to each preset place type to determine a second number of addresses in the preset address, which are matched with the name words.
And determining the corresponding preset place types of which the first quantity is not less than a first preset threshold value and the second quantity is not less than a second preset threshold value as the preset place types corresponding to the preset WIFI network.
And establishing a corresponding relation between the preset WIFI identification and the preset place type according to the WIFI identification of the preset WIFI network and the preset place type corresponding to the preset WIFI network.
Optionally, the second obtaining module 302 is further specifically configured to:
and aiming at the preset place type, acquiring a WIFI network of which the set place belongs to the preset place type as a sample WIFI network.
And acquiring a WIFI address of the sample WIFI network to serve as a preset address corresponding to the preset place type, and performing word segmentation processing on the WIFI name of the sample WIFI network to obtain a sample word.
And selecting preset words from the sample words according to the occurrence times of the sample words so as to generate the preset word packet.
Optionally, the second obtaining module 302 is further specifically configured to:
and calculating the absolute value of the difference value of the occurrence parameter and the preset parameter threshold value.
Determining the second interest value according to the absolute value; wherein the second interest value is negatively correlated with the absolute value in the event that the occurrence parameter is not greater than the preset parameter threshold; the second interest value is positively correlated with the absolute value if the occurrence parameter is greater than the preset parameter threshold.
Optionally, the first obtaining module 301 is specifically configured to:
taking the user characteristics as the input of the preset classification model, and generating a first interest value by using the preset classification model; the user characteristics are determined according to the personal information of the user to be determined.
Or searching a first interest value corresponding to the user ID from a preset first interest value database according to the user identification ID of the user to be determined to obtain the first interest value of the user to be determined to the preset content category; the preset first interest value database stores first interest values of different users for the preset content categories, and the first interest values stored in the first interest value database are generated in advance according to the preset classification model and the user characteristics of the users.
In summary, the parameter obtaining apparatus provided in the embodiment of the present invention may obtain, for a user to be determined, a first interest value of the user to be determined in a preset content category, where the first interest value is determined by using a user characteristic of the user to be determined and a preset classification model, the user characteristic is determined according to user personal information of the user to be determined, then, a second interest value of the user to be determined in the preset content category may be obtained according to a place where the user to be determined appears in a preset time period and a preset place type corresponding to the preset content category, and finally, a final interest value of the user to be determined in the preset content category is calculated according to the first interest value and the second interest value. The method and the device have the advantages that the position where the user to be determined appears in the preset time length can reflect the nearest short-term interest of the user to be determined to a certain extent, so that the method and the device can determine the final interest value reflecting the interest of the user by using the first interest value determined according to the personal information of the user and combining the second interest value determined according to the occurrence parameter, the accuracy of the determined final interest value reflecting the interest of the user can be improved to a certain extent, and therefore when the user is recommended based on the final interest value in the following process, the recommended content to the user can meet the requirements of the user better, and the recommendation effect is improved.
An embodiment of the present invention further provides an electronic device, as shown in fig. 4, including a processor 401, a communication interface 402, a memory 403, and a communication bus 404, where the processor 401, the communication interface 402, and the memory 403 complete mutual communication through the communication bus 404,
a memory 403 for storing a computer program;
the processor 401, when executing the program stored in the memory 403, implements the following steps:
for a user to be determined, acquiring a first interest value of the user to be determined in a preset content category; the first interest value is determined by using the user characteristics of the user to be determined and a preset classification model.
And acquiring a second interest value of the user to be determined on the preset content category according to the place where the user to be determined appears in the preset time period and the preset place type corresponding to the preset content category.
And calculating the final interest value of the user to be determined in the preset content category according to the first interest value and the second interest value.
The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the terminal and other equipment.
The Memory may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
In another embodiment of the present invention, a computer-readable storage medium is further provided, in which instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the parameter obtaining method described in any one of the above embodiments.
In yet another embodiment of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the parameter acquisition method as described in any of the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (11)
1. A method for parameter acquisition, the method comprising:
for a user to be determined, acquiring a first interest value of the user to be determined in a preset content category; the first interest value is determined by utilizing the user characteristics of the user to be determined and a preset classification model;
acquiring a second interest value of the user to be determined on the preset content category according to the place where the user to be determined appears in the preset time period and the preset place type corresponding to the preset content category;
and calculating the final interest value of the user to be determined in the preset content category according to the first interest value and the second interest value.
2. The method according to claim 1, wherein after calculating a final interest value of the user to be determined in the preset content category according to the first interest value and the second interest value, the method further comprises:
and acquiring the recommendation information of the user to be determined according to the final interest value of the user to be determined in the preset content category, and outputting the recommendation information to the user to be determined.
3. The method according to claim 1 or 2, wherein the obtaining of the second interest value of the user to be determined in the preset content category according to the place where the user to be determined has appeared in the preset time period and the preset place type corresponding to the preset content category includes:
acquiring occurrence parameters of the to-be-determined user in a place of a target preset place type according to the place where the to-be-determined user appears in a preset time period; the target preset place type is a preset place type corresponding to the preset content type;
and calculating a second interest value of the user to be determined in the preset content category according to the occurrence parameter and a preset parameter threshold corresponding to the type of the target preset place.
4. The method according to claim 3, wherein the obtaining of the occurrence parameter of the to-be-determined user at the location of the target preset location type according to the location where the to-be-determined user has occurred within the preset time period comprises:
acquiring a WIFI identifier of a historical wireless local area network WIFI network connected by the user to be determined within the preset time period;
dividing the historical WIFI network into WIFI networks which are matched with the historical WIFI network and are included in preset place types according to the corresponding relation among the WIFI identifications, the preset WIFI identifications and the preset place types;
and counting the sum of the connection times and/or the connection duration of the user to be determined and the WIFI network included in the target preset place type to serve as the occurrence parameter.
5. The method of claim 4, wherein the WIFI identifier comprises a WIFI name and a WIFI address; before dividing the historical WIFI network into WIFI networks included in preset location types matched with the historical WIFI network according to the corresponding relation among the WIFI identifiers, the preset WIFI identifiers and the preset location types, the method further comprises the following steps:
for the preset WIFI network, segmenting the WIFI name of the preset WIFI network to obtain a name word;
matching the name words with preset word packets corresponding to each preset location type to determine a first number of words in the preset word packets, which are matched with the name words, and matching the WIFI address of the preset WIFI network with a preset address corresponding to each preset location type to determine a second number of addresses in the preset address, which are matched with the name words;
determining the corresponding preset location types of which the first quantity is not less than a first preset threshold value and the second quantity is not less than a second preset threshold value as the preset location types corresponding to the preset WIFI network;
and establishing a corresponding relation between the preset WIFI identification and the preset place type according to the WIFI identification of the preset WIFI network and the preset place type corresponding to the preset WIFI network.
6. The method of claim 5, wherein before matching the name term with the preset bundle of terms corresponding to each preset locality type, the method further comprises:
acquiring a WIFI network of which the set place belongs to the preset place type as a sample WIFI network aiming at the preset place type;
acquiring a WIFI address of the sample WIFI network to serve as a preset address corresponding to the preset place type, and performing word segmentation processing on the WIFI name of the sample WIFI network to obtain a sample word;
and selecting preset words from the sample words according to the occurrence times of the sample words so as to generate the preset word packet.
7. The method according to claim 3, wherein the calculating a second interest value of the user to be determined in the preset content category according to the occurrence parameter and a preset parameter threshold corresponding to the type of the target preset location includes:
calculating the absolute value of the difference value of the occurrence parameter and the preset parameter threshold;
determining the second interest value according to the absolute value; wherein the second interest value is negatively correlated with the absolute value in the event that the occurrence parameter is not greater than the preset parameter threshold; the second interest value is positively correlated with the absolute value if the occurrence parameter is greater than the preset parameter threshold.
8. The method according to claim 1, wherein the obtaining of the first interest value of the user to be determined in the preset content category comprises:
taking the user characteristics as the input of the preset classification model, and generating a first interest value by using the preset classification model; the user characteristics are determined according to personal information of the user to be determined;
or searching a first interest value corresponding to the user ID from a preset first interest value database according to the user identification ID of the user to be determined to obtain the first interest value of the user to be determined to the preset content category; the preset first interest value database stores first interest values of different users for the preset content categories, and the first interest values stored in the first interest value database are generated in advance according to the preset classification model and the user characteristics of the users.
9. An apparatus for parameter acquisition, the apparatus comprising:
the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a first interest value of a user to be determined on a preset content category; the first interest value is determined by utilizing the user characteristics of the user to be determined and a preset classification model;
the second obtaining module is used for obtaining a second interest value of the user to be determined on the preset content category according to the place where the user to be determined appears in the preset time period and the preset place type corresponding to the preset content category;
and the calculating module is used for calculating the final interest value of the user to be determined in the preset content category according to the first interest value and the second interest value.
10. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;
a memory for storing a computer program;
a processor for implementing the method steps of any of claims 1 to 8 when executing a program stored in the memory.
11. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010340907.8A CN111695030A (en) | 2020-04-26 | 2020-04-26 | Parameter acquisition method and device, electronic equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010340907.8A CN111695030A (en) | 2020-04-26 | 2020-04-26 | Parameter acquisition method and device, electronic equipment and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111695030A true CN111695030A (en) | 2020-09-22 |
Family
ID=72476637
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010340907.8A Pending CN111695030A (en) | 2020-04-26 | 2020-04-26 | Parameter acquisition method and device, electronic equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111695030A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110060649A1 (en) * | 2008-04-11 | 2011-03-10 | Dunk Craig A | Systems, methods and apparatus for providing media content |
| US20180075512A1 (en) * | 2016-09-13 | 2018-03-15 | Adobe Systems Incorporated | Item recommendation techniques |
| CN107820209A (en) * | 2017-11-07 | 2018-03-20 | 锐捷网络股份有限公司 | Interest recommends method, apparatus and server |
| CN109041035A (en) * | 2018-09-28 | 2018-12-18 | Oppo广东移动通信有限公司 | User's portrait generation method, device, terminal and storage medium |
| CN109670104A (en) * | 2018-11-12 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Information-pushing method, unit and storage medium based on machine learning |
-
2020
- 2020-04-26 CN CN202010340907.8A patent/CN111695030A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110060649A1 (en) * | 2008-04-11 | 2011-03-10 | Dunk Craig A | Systems, methods and apparatus for providing media content |
| US20180075512A1 (en) * | 2016-09-13 | 2018-03-15 | Adobe Systems Incorporated | Item recommendation techniques |
| CN107820209A (en) * | 2017-11-07 | 2018-03-20 | 锐捷网络股份有限公司 | Interest recommends method, apparatus and server |
| CN109041035A (en) * | 2018-09-28 | 2018-12-18 | Oppo广东移动通信有限公司 | User's portrait generation method, device, terminal and storage medium |
| CN109670104A (en) * | 2018-11-12 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Information-pushing method, unit and storage medium based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106897428B (en) | Text classification feature extraction method and text classification method and device | |
| CN108073568B (en) | Keyword extraction method and device | |
| CN107862022B (en) | Culture resource recommendation system | |
| WO2020207074A1 (en) | Information pushing method and device | |
| CN109189990B (en) | Search word generation method and device and electronic equipment | |
| CN109165691B (en) | Training method and device for model for identifying cheating users and electronic equipment | |
| CN109165975B (en) | Label recommending method, device, computer equipment and storage medium | |
| CN110689368B (en) | Method for designing advertisement click rate prediction system in mobile application | |
| CN109190014B (en) | Regular expression generation method and device and electronic equipment | |
| CN112488716B (en) | Abnormal event detection system | |
| US20160170993A1 (en) | System and method for ranking news feeds | |
| CN111159563A (en) | Method, device and equipment for determining user interest point information and storage medium | |
| CN116881429B (en) | Multi-tenant-based dialogue model interaction method, device and storage medium | |
| CN108319628B (en) | User interest determination method and device | |
| CN111639696A (en) | User classification method and device | |
| CN114254615A (en) | Volume assembling method and device, electronic equipment and storage medium | |
| CN112269906B (en) | Automatic extraction method and device of webpage text | |
| CN111930949B (en) | Search string processing method and device, computer readable medium and electronic equipment | |
| CN113220974A (en) | Click rate prediction model training and search recall method, device, equipment and medium | |
| US10346856B1 (en) | Personality aggregation and web browsing | |
| CN115329078B (en) | Text data processing method, device, equipment and storage medium | |
| CN112685618A (en) | User feature identification method and device, computing equipment and computer storage medium | |
| CN111695030A (en) | Parameter acquisition method and device, electronic equipment and computer readable storage medium | |
| CN113962216A (en) | Text processing method and device, electronic equipment and readable storage medium | |
| CN114547440A (en) | User portrait mining method based on internet big data and artificial intelligence cloud system |
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 |