CN105488233A - Reading information recommendation method and system - Google Patents

Abstract

The present invention relates to a method and system for recommending reading information, wherein the method includes the following steps: obtaining historical reading data of a target user; wherein the historical reading data includes content information of the target user accessing a reading website through an intelligent terminal Obtaining the interest information of the target user according to the historical reading data; constructing a similar user group according to the interest information of the target user to collect information related to the interest information of the target user in the reading website other users, and obtain the interest information of other users in the similar user group; generate recommendation information of the target user according to the interest information of the target user and the interest information of other users; recommend the recommendation information to the Target users. Through the above technical solutions, the reading information recommendation method and system of the present invention not only improves the accuracy of the recommended reading information, but also realizes the diversity of the recommended reading information.

Description

Reading information recommendation method and system

technical field

The invention relates to the technical field of the Internet, in particular to a reading information recommendation method and system.

Background technique

With the rapid development of Internet technology, the information on the Internet is also showing explosive growth. The scale of digital information in today's society is getting larger and larger, and the ways to obtain information are becoming wider and wider. People can always find information on the Internet. All kinds of information resources you need.

However, information overload also prevents us from efficiently obtaining the information and resources we need most from the Internet. In the context of the era of big data, the way people obtain information has undergone tremendous changes. At present, most push technologies are mostly based on the similarity of products or information. The disadvantage of this development is that more and more users will focus on popular products or hot search information, while those that are not popular may be of interest to users. Products or information will be ignored.

Reading e-books has increasingly become a mainstream reading method under the tide of the Internet. People obtain the information they need by reading the reading resources recommended on the website. E-books have won many reading lovers for their portability, easy access and immediacy. of favor. However, traditional e-reading websites mostly recommend reading content to users based on the historical reading information of users who visit the reading website, resulting in that the content of the recommended information received by the user is relatively single, which cannot meet the needs of diversity.

Contents of the invention

Based on this, it is necessary to provide a reading information recommendation method and system for the technical problem that the content of the recommended information in the existing information recommendation method is relatively single and cannot meet the requirements of diversity.

A method for recommending reading information, comprising the steps of:

Acquiring the historical reading data of the target user; wherein, the historical reading data includes the content information of the target user accessing the reading website through the smart terminal;

acquiring interest information of the target user according to the historical reading data;

Constructing a similar user group based on the interest information of the target user, which is used to collect other users related to the interest information of the target user in the reading website, and obtain the interest information of other users in the similar user group ;

generating recommendation information of the target user according to the interest information of the target user and the interest information of other users;

Recommending the recommendation information to the target user.

A reading information recommendation system, comprising:

The first obtaining module is used to obtain the historical reading data of the target user; wherein, the historical reading data includes the content information of the target user accessing the reading website through the smart terminal;

A second acquiring module, configured to acquire the interest information of the target user according to the historical reading data;

A building module, configured to construct a similar user group based on the interest information of the target user, to gather other users in the reading website that are related to the interest information of the target user, and to acquire other users in the similar user group. User's interest information;

A generating module, configured to generate recommendation information of the target user according to the interest information of the target user and other user interest information;

A recommendation module, configured to recommend the recommendation information to the target user.

The above reading information recommendation method and system, by combining the interest information of the target user with the interest information of other users in similar user groups, not only recommends the information that the target user is interested in, but also improves the recommendation efficiency. The accuracy of the reading information; moreover, the information of interest of other users in the similar user group is also recommended to the target user, realizing the diversity of the recommended reading information.

Description of drawings

Fig. 1 is a flowchart of a reading information recommendation method according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a reading information recommendation system according to an embodiment of the present invention.

detailed description

In order to further illustrate the technical means adopted by the present invention and the achieved effects, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and preferred embodiments.

As shown in Figure 1, Figure 1 is a flowchart of a method for recommending reading information according to an embodiment of the present invention, including the following steps:

Step S101: Obtain historical reading data of the target user; wherein, the historical reading data includes the content information of the target user's access to the reading website through a smart terminal;

In this step, according to the content information that the target user accesses the reading website through the smart terminal, for example, which type of information in which reading website the target user accesses through the smart terminal, and what content has been read, as well as comments or concerns What content information and so on.

Step S102: Obtain the interest information of the target user according to the historical reading data;

In this step, the interest information of the target user is acquired according to the above historical reading data. The interest information of the target user described in the present invention includes the reading websites that the target user frequently visits through the smart terminal, which type of articles the target user frequently reads on a certain reading website, and so on.

Step S103: Construct a similar user group according to the interest information of the target user, which is used to collect other users related to the interest information of the target user in the reading website, and obtain the information of other users in the similar user group. information of interest;

In this step, other users who have the same interest information as the target user are gathered by constructing a similar user group. Of course, these users in the similar user group not only have the same interest information as the target user, but also have their own interest information. By obtaining the interest information of other users in the similar user group, effectively Expand the range of information that the target user is interested in.

Step S1014: Generate recommendation information of the target user according to the target user's interest information and other user's interest information;

In this step, considering that the target user may not only have the same interest information as the target user, but also have consistent interest information with that of other users, in order to expand the selection of recommended information for the target user Scope, combining the interest information of the target user with the interest information of other users to generate the final recommendation information of the target user. Through the above operations, the reading information recommendation method of the present invention not only recommends the target user's interest information to the target user, and improves the accuracy of the recommended information; Interested information is recommended to target users, expanding the selection range of target users' recommended information, and realizing the diversity of recommended information.

Step S105: Recommend the recommendation information to the target user.

In this step, according to the recommendation information generated above, the recommendation information is recommended to target users, so that the target users can read the information they are interested in, and can also read other reading information related to themselves, Broaden the range of reading information and enhance the diversity of reading information.

The above reading information recommendation method, by combining the interest information of the target user with the interest information of other users in similar user groups, not only recommends the information that the target user is interested in, but also improves the level of the recommended reading information. accuracy; moreover, the information of interest of other users in the similar user group is also recommended to the target user, realizing the diversity of the recommended reading information.

In practical applications, reading is of course the most important requirement for users, but if reading information is pushed to users when they are sleeping, the recommendation effect may be counterproductive, so it is not only necessary to know what reading information users like , but also need to know the user's reading behavior information, so as to provide personalized recommendation services.

In one of the embodiments, the reading information recommendation method of the present invention may further include: obtaining historical behavior information of the target user visiting the reading website through the smart terminal;

The reading behavior characteristics of the target user are acquired according to the historical behavior information, and the recommendation information is recommended to the target user according to the reading behavior characteristics.

The historical behavior information described in the above embodiments refers to the behavior habit of the target user accessing the reading website through the smart terminal, for example, which type of reading website is visited in what time period, and which type of reading website is read in which time period. What kind of books, and the reading cycle in various time periods, etc. Acquire the reading behavior characteristics of the target user according to the historical behavior information, that is, the reading time period, historical reading times, reading activity, etc. of the target user within a preset time period. After the above-mentioned reading behavior characteristics of the target user are acquired, when recommending reading information to the target user, the reading information may be recommended in combination with the above-mentioned reading behavior characteristics. Through the above operations, the personalized recommendation to the target user is realized.

In practical applications, the user's personalized recommendation list can be generated by screening the data in three dimensions: the user's weekly interest preference, historical daily reading dominance time preference, and user's reading and visiting website preference.

Weekly interest preference: Screen the personalized recommendation results to select books that match the user's weekly interest data, and use the book with the highest resource score in the personalized recommendation results as the user's recommended book.

Historical daily reading dominance time preference: formulate personalized recommendation items for books that are pushed regularly.

User preference to visit websites: According to the user's preference for website visits, make personalized push items for e-reading websites with the bibliography.

In one of the embodiments, after the step of acquiring the historical reading data of the target user, the method for recommending reading information of the present invention may further include:

Perform privacy transcoding and integrity processing on the historical reading data.

The reading information recommendation method above protects the reading privacy information of the target user to a certain extent by performing privacy transcoding processing on the acquired historical reading data; by performing integrity processing on the acquired historical reading data, the acquired historical reading data A more accurate and comprehensive reflection of the target user's historical reading behavior is conducive to improving the accuracy of the recommended reading information for the target user.

In one of the embodiments, in the reading information recommendation method of the present invention, the information of interest may include categories of books of interest;

The step of obtaining the interest information of the target user according to the content information includes:

According to the content information, obtain the category of books that the target user is interested in;

The step of generating the recommendation information of the target user according to the interest information of the target user and the interest information of other users, and recommending the recommendation information to the target user includes:

generating book category recommendation information for the target user according to the book category of interest to the target user and the book categories of interest to other users;

The step of recommending the recommendation information to the target user includes:

Recommending the book category recommendation information to the target user.

In the above embodiments, the reading information recommendation method of the present invention obtains the book categories that the target user is interested in and the book categories that other users are interested in, and based on the book categories that the target user is interested in and the books that other users are interested in The category generates book category recommendation information for the target user, and recommends the book category recommendation information to the target user. Through the above scheme, not only recommend the book categories that the target user is interested in, and improve the accuracy of the recommended reading information; moreover, the book categories that other users who have the same interest information as the target user are also recommended to the target user, Realized the diversity of recommended reading information.

In practical applications, the categories of books mentioned in the information recommendation method of the present invention include but are not limited to: urban romance, fantasy, fantasy, martial arts, game competition, romantic youth, historical military, supernatural suspense, famous biography, Science fiction, film and television scripts, fashion life, officialdom and workplace, economics and management inspiration, educational social sciences and short stories.

In one embodiment, in the reading information recommendation method of the present invention, the step of recommending the book category recommendation information to the target user may include:

According to the book category recommendation information of the target user, the books matching the book category recommendation information are screened out, and sorted according to the matching degree to generate a reading book recommendation list;

According to the reading book recommendation list, reading books are recommended to the target user.

In the above embodiment, the reading information recommendation method of the present invention generates a reading book recommendation list by filtering out books matching the book category recommendation information and sorting them according to the degree of matching. When recommending books to the target user, books are recommended to the target user in order of matching degree from high to low, realizing personalized and precise recommendation to the target user.

In practical applications, it is also possible to filter books that match the recommended book category information by extracting data keywords. The specific operations are as follows:

(1) Extract data keywords, perform data matching on the "keyword" field in the book information data of the reading website according to the most basic categories of text vocabulary, and perform statistics on the number of occurrences to obtain the corresponding list of books of each different category "Keywords" field.

(2) Arrange the matched "keywords", filter out the top keywords with the highest frequency, and use them as the screening keywords for books of this category.

(3) Formulate screening rules: 1) If the title of a book contains screening keywords, it is considered to be a book of this category. 2) If the number of screening keywords in the content introduction of the book is greater than a certain amount, it is considered to be a book of this category.

After the above data processing, the target user's historical reading data and reading behavior information are obtained by analyzing the url access data of the target user's access to the reading website through the smart terminal. When obtaining the interested information of the target user, the user's access url is parsed into the historical reading data of reading; through the five behavioral index data of daily reading domination time, weekly reading domination time, activity, explosiveness and memory, to Describe the reading behavior information of target users.

In one of the embodiments, the reading information recommendation method of the present invention, before the step of recommending the recommendation information to the target user, further includes:

Determining the identity feature information of the target user;

The step of recommending the recommendation information to the target user includes:

According to the identity feature information, recommend information to the target user.

In the above embodiment, before recommending reading information to the target user, the reading information recommendation method of the present invention may also determine the identity feature information of the target user, for example, the industry the target user is engaged in, the level he is in, etc. etc., and then recommend information to the target user according to the identity feature information, further realizing personalized and precise recommendation to the target user.

In practical applications, in order to meet the needs of smart terminal reading product operators in terms of marketing business, before making personalized recommendations to users, the behavior or identity characteristics of users will be screened, and a category with clear characteristics will be selected from them. users, and make personalized recommendations for them. The specific screening rules are as follows:

Screen users with the same occupation;

Screen users who read the same dominant time period every day by setting a time period;

Screen users with the same dominant reading time period every week by setting a time period;

Filter users whose historical reading times are greater or less than the threshold by setting a threshold;

Filter users whose activity is greater or less than the threshold in historical reading records by setting a threshold;

By setting a threshold, count and select users whose activity in the previous month's reading records is greater than or less than the threshold.

By using the above screening rules, personalized and accurate recommendations for specific user groups can be realized.

In practical application, the reading information recommendation method of the present invention can also grab target user's behavior data from the web through web crawlers, and the data sources are mainly mobile phone reading websites and similar reading websites. Among them, behavioral data includes:

① Book information data. Book title, book ID, author, category name, category ID, cover image, introduction, price (including physical book price, e-book price, purchase price by chapter, reading package coverage), word count, chapter count, keywords, home page link, first chapter link and website name.

②Evaluation index data. Indicator name, indicator measurement effect description, indicator calculation method and indicator calculation data caliber.

③ Read behavioral data. User ID, date, reading PV, book ID.

④ Read the recorded data. User reading historical PV records, user login, payment/monthly subscription, subscription, and client use.

⑤ User bills and basic information. The user's ARPU value, user basic information (gender, birthday, occupation), user call/sms details, user value-added service usage.

⑥The user accesses the url data. url address, access time, return data packet length and access location.

According to the above behavior data, reading information can be recommended to target users in a targeted manner, and personalized recommendation of reading information can be realized.

In practical applications, the reading information recommendation method of the present invention can also use the recommendation algorithm on the above-mentioned behavior data of the target user, and extract all the items whose resource scores are not 0, and which have not been selected by the user. Sort in descending order according to the resource score, that is, generate a personalized recommendation list for the user.

As shown in Figure 2, Figure 2 is a schematic structural diagram of a reading information recommendation system according to an embodiment of the present invention, including:

The first obtaining module 101 is used to obtain the historical reading data of the target user; wherein, the historical reading data includes the content information of the target user's access to the reading website through the smart terminal;

In this embodiment, the first acquisition module 101 of the reading information recommendation system of the present invention is based on the content information of the reading website accessed by the target user through the smart terminal, for example, which of the reading websites the target user visits through the smart terminal Class information, and what content has been read, and what content information has been commented on or followed, and so on.

The second acquiring module 102 is configured to acquire the interest information of the target user according to the historical reading data;

In this embodiment, the second acquisition module 102 of the reading information recommendation system of the present invention acquires the interest information of the target user according to the historical reading data as described above. The interest information of the target user described in the present invention includes the reading websites that the target user frequently visits through the smart terminal, which type of articles the target user frequently reads on a certain reading website, and so on.

A construction module 103, configured to construct a similar user group according to the interest information of the target user, to gather other users in the reading website that are related to the interest information of the target user, and to obtain information in the similar user group Information of interest to other users;

In this embodiment, the construction module 103 of the reading information recommendation system of the present invention gathers other users who have the same information of interest as the target user by building a similar user group. Of course, these users in the similar user group not only have the same interest information as the target user, but also have their own interest information. By obtaining the interest information of other users in the similar user group, effectively Expand the range of information that the target user is interested in.

A generation module 104, configured to generate recommendation information of the target user according to the interest information of the target user and the interest information of other users;

In this embodiment, considering that the target user may not only have the same interest information as the target user, but also have the same interest information as that of other users, in order to expand the target user recommendation information Select the scope, the generation module 104 of the reading information recommendation system of the present invention combines the interest information of the target user with the interest information of other users to generate the final recommendation information of the target user. Through the above operations, the reading information recommendation system of the present invention not only recommends the target user's interest information to the target user, and improves the accuracy of the recommended information; Interested information is recommended to target users, expanding the selection range of target users' recommended information, and realizing the diversity of recommended information.

A recommendation module 105, configured to recommend the recommendation information to the target user.

The recommendation module 105 of the reading information recommendation system of the present invention is based on the recommendation information generated above, and recommends the recommendation information to target users, so that the target users can read the information they are interested in, and can also read information related to themselves. Other relevant reading information broadens the scope of reading information and enhances the diversity of reading information.

The above-mentioned reading information recommendation system not only recommends the information that the target user is interested in by combining the interest information of the target user with the interest information of other users in the similar user group, but also improves the level of the recommended reading information. accuracy; moreover, the information of interest of other users in the similar user group is also recommended to the target user, realizing the diversity of the recommended reading information.

In one of the embodiments, in the reading information recommendation system of the present invention, the first acquisition module 101 can also be used to: acquire the behavior information of the target user visiting the reading website through a smart terminal;

The second acquisition module 102 may also be used to: acquire the reading behavior characteristics of the target user according to the behavior information;

The recommendation module 104 may also be configured to: recommend the recommendation information to the target user according to the reading behavior characteristics.

The historical behavior information described in the above embodiments refers to the behavior habit of the target user accessing the reading website through the smart terminal, for example, which type of reading website is visited in what time period, and which type of reading website is read in which time period. What kind of books, and the reading cycle in various time periods, etc. Acquire the reading behavior characteristics of the target user according to the historical behavior information, that is, the reading time period, historical reading times, reading activity, etc. of the target user within a preset time period. After the above-mentioned reading behavior characteristics of the target user are acquired, when recommending reading information to the target user, the reading information may be recommended in combination with the above-mentioned reading behavior characteristics. Through the above operations, the personalized recommendation to the target user is realized.

In one of the embodiments, in the reading information recommendation system of the present invention, the reading behavior characteristics of the target user include:

The reading time period, historical reading times and reading activity of the target user within the preset time period.

In one of the embodiments, the reading information recommendation system of the present invention, after the first acquisition module 101, further includes:

A preprocessing module is used to perform privacy transcoding and integrity processing on the historical reading data.

The above-mentioned reading information recommendation system, through the preprocessing module, performs privacy transcoding processing on the acquired historical reading data, which protects the reading privacy information of the target user to a certain extent; Integrity processing enables the acquired historical reading data to reflect the target user's historical reading behavior more accurately and comprehensively, which is conducive to improving the accuracy of recommended reading information for the target user.

In one of the embodiments, in the reading information recommendation system of the present invention, the information of interest includes the category of books of interest;

The second obtaining module 102 can also be used to obtain the book category that the target user is interested in according to the content information;

The generating module 103 can also be used to generate book category recommendation information for the target user according to the book category of interest to the target user and the book categories of interest to other users;

The recommendation module 105 can also be used to recommend the book category recommendation information to the target user.

In the above-mentioned embodiments, the reading information recommendation system of the present invention obtains the book categories that the target user is interested in and the book categories that other users are interested in, and based on the book categories that the target user is interested in and the books that other users are interested in The category generates book category recommendation information for the target user, and recommends the book category recommendation information to the target user. Through the above scheme, the reading information recommendation system of the present invention not only recommends the book categories that the target user is interested in, and improves the accuracy of the recommended reading information; Book categories are recommended to target users, realizing the diversity of recommended reading information.

In one embodiment, in the reading information recommendation system of the present invention, the recommendation module 105 can also be used to: filter out books matching the book category recommendation information according to the target user's book category recommendation information, And sort according to the matching degree to generate a recommended list of reading books;

According to the reading book recommendation list, reading books are recommended to the target user.

In the above embodiment, the reading information recommendation system of the present invention generates a reading book recommendation list by filtering out books that match the book category recommendation information and sorting them according to the degree of matching. When recommending books to the target user, books are recommended to the target user in order of matching degree from high to low, realizing personalized and precise recommendation to the target user.

In one of the embodiments, before the recommendation module 105, the reading information recommendation system of the present invention may further include: an identity determination module, configured to determine the identity feature information of the target user;

The recommendation module 105 may also be configured to: recommend information to the target user according to the identity feature information.

In the above embodiment, before recommending reading information to the target user, the reading information recommendation system of the present invention may also determine the identity feature information of the target user through the identity determination module, for example, the industry the target user is engaged in, the The level of the location, etc., and then recommend information to the target user according to the identity feature information, and further realize the personalized and precise recommendation to the target user.

The above-mentioned reading information recommendation system not only recommends the information that the target user is interested in by combining the interest information of the target user with the interest information of other users in the similar user group, but also improves the level of the recommended reading information. accuracy; moreover, the information of interest of other users in the similar user group is also recommended to the target user, realizing the diversity of the recommended reading information.

The various technical features of the above-mentioned embodiments can be combined arbitrarily. For the sake of concise description, all possible combinations of the various technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, should be considered as within the scope of this specification.

The above-mentioned embodiments only express several implementation modes of the present invention, and the descriptions thereof are relatively specific and detailed, but should not be construed as limiting the patent scope of the invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention should be based on the appended claims.

Claims (10)

1. A reading information recommendation method, characterized in that, comprising the steps of: Acquiring the historical reading data of the target user; wherein, the historical reading data includes the content information of the target user accessing the reading website through the smart terminal; acquiring interest information of the target user according to the historical reading data; Constructing a similar user group based on the interest information of the target user, which is used to collect other users related to the interest information of the target user in the reading website, and obtain the interest information of other users in the similar user group ; generating recommendation information of the target user according to the interest information of the target user and the interest information of other users; Recommending the recommendation information to the target user. 2. The method for recommending reading information according to claim 1, further comprising: acquiring historical behavior information of the target user accessing the reading website through a smart terminal; The reading behavior characteristics of the target user are acquired according to the historical behavior information, and the recommendation information is recommended to the target user according to the reading behavior characteristics. 3. The reading information recommendation method according to claim 2, wherein the reading behavior characteristics of the target user include: The reading time period, historical reading times and reading activity of the target user within the preset time period. 4. The method for recommending reading information according to claim 1, further comprising: Perform privacy transcoding and integrity preprocessing on the historical reading data. 5. The method for recommending reading information according to claim 1, wherein the information of interest includes categories of books of interest; The step of obtaining the interest information of the target user according to the content information includes: According to the content information, obtain the category of books that the target user is interested in; The step of generating the recommendation information of the target user according to the interest information of the target user and the interest information of other users includes: Generate book category recommendation information for the target user according to the book category of interest to the target user and the book categories of interest to other users; The step of recommending the recommendation information to the target user includes: Recommending the book category recommendation information to the target user. 6. The reading information recommendation method according to claim 5, wherein the step of recommending the book category recommendation information to the target user comprises: According to the book category recommendation information of the target user, the books matching the book category recommendation information are screened out, and sorted according to the matching degree to generate a reading book recommendation list; According to the reading book recommendation list, reading books are recommended to the target user. 7. The method for recommending reading information according to claim 1, further comprising: before the step of recommending the recommended information to the target user: determining the identity feature information of the target user; The step of recommending the recommendation information to the target user includes: According to the identity feature information, recommend information to the target user. 8. A reading information recommendation system, characterized in that, comprising: The first obtaining module is used to obtain the historical reading data of the target user; wherein, the historical reading data includes the content information of the target user accessing the reading website through the smart terminal; A second acquiring module, configured to acquire the interest information of the target user according to the historical reading data; A building module, configured to construct a similar user group based on the interest information of the target user, to gather other users in the reading website that are related to the interest information of the target user, and to acquire other users in the similar user group. User's interest information; A generating module, configured to generate recommendation information of the target user according to the target user's interest information and other user's interest information; A recommendation module, configured to recommend the recommendation information to the target user. 9. The reading information recommendation system according to claim 8, wherein the first acquiring module is further used to: acquire the historical behavior information of the target user accessing the reading website through an intelligent terminal; The second acquisition module is further configured to: acquire the reading behavior characteristics of the target user according to the historical behavior information. 10. The reading information recommendation system according to claim 9, wherein the recommendation module is also used for: Recommending the recommendation information to the target user according to the reading behavior characteristics.