Personalized Recommendation Preview Method Based on Web Search

Abstract

A personalized recommendation preview method based on web search provides a first display area and a second display area to form a search page. Among them, the first display area is used to display web pages searched based on the search information typed by the user, and the second display area is used to display web pages recommended to the user after helping the user perform associated searches based on the search information typed by the user. The present invention first roughly screens out recommended users for the current user through the browsing duration and browsing times of the user on the web pages. Then, on the basis of the browsing duration and browsing times, it further considers the user's operation conditions on the web pages, and recommends web pages that best meet the user's needs to the user through more refined screening indicators, greatly improving the accuracy of the recommendation.

Description

Technical Field

[0001] The present invention relates to the technical field of web search recommendation, and particularly to a personalized recommendation preview method based on web search. Background Art

[0002] The rapid development of the Internet and the increase in a vast amount of network resources have brought many conveniences to users' work and life. How to accurately and comprehensively obtain the information required by users from the vast amount of network resources is the goal that major search engines strive to achieve. Existing web searches provide various algorithms to provide users with the required resources. Most of these algorithms aim to improve the accuracy of searches based on keywords and users' historical search information. However, the existing search methods currently only target the user's current search needs in a single manner, and the provided web pages only involve a single search need. However, in different scenarios such as actual life, work, and study, when users search for network resources each time, there are usually not only one search need. Often, based on the development of the search work, other related searches are required. If the targeted search for the current keyword and the related search based on the current keyword can be provided to users simultaneously, it can greatly improve the user-friendliness of the search page and also help users greatly improve the search efficiency. Summary of the Invention

[0003] To solve the deficiencies of the existing technology, the present invention provides a personalized recommendation preview method based on web search, including the following steps:

[0004] Step S1: Obtain keywords from the search information input by the current user, perform a web search based on the keywords, and display the searched web pages in the first display area of the search page;

[0005] Step S2: Determine the web page label, where the web page label is the category to which the web page belongs;

[0006] Step S3: Obtain the current user's web page set, where the current user's web page set is the set of all web pages browsed by the current user within a first predetermined time period and related to the web page label determined in Step S2;

[0007] Step S4: Determine the associated users based on the current user's web page set; where the associated users are the users whose sets of all web pages browsed within a second predetermined time period and related to the web page label determined in Step S2 overlap with the current user's web page set;

[0008] Step S5: Determine the operation similarity between the current user and the associated users based on the historical operation set of the current user and the historical operation set of the associated users;

[0009] Step S6: Screen and recommend users from the associated users based on the operation similarity;

[0010] Step S7: Generate a recommended web page based on the historical operation set of the current user and the historical operation set of the recommended user, and display the recommended web page in the second display area of the search page.

[0011] Among them, the historical operation set of the current user includes each web page in the current user's web page set and the web page weight corresponding to each web page;

[0012] The historical operation set of the associated user includes each web page in the associated user's web page set and the web page weight corresponding to each web page;

[0013] The historical operation set of the recommended user includes each web page in the recommended user's web page set and the web page weight corresponding to each web page;

[0014] The associated user's web page set is all web pages related to the web page tags determined in step S2 that the associated user browsed within the second predetermined time period;

[0015] The recommended user's web page set is all web pages related to the web page tags determined in step S2 that the recommended user browsed within the second predetermined time period;

[0016] The web page weight is determined by the browsing duration and the number of browsing times.

[0017] Among them, the operation similarity between the current user and the associated user is determined by the following formula:

[0018] ;

[0019] Among them, W represents the operation similarity between the current user and the associated user;

[0020] A represents the current user's web page set, including web pages a1, a2, a3... a n , and the web page weights of web pages a1, a2, a3... a n are λ1, λ2, λ3... λ n ;

[0021] B represents the associated user's web page set, including web pages b1, b2, b3... b m , and the web page weights of web pages b1, b2, b3... b m are β1, β2, β3... β m ;

[0022] n and m respectively represent the number of web pages in the current user's web page set and the associated user's web page set;

[0023] A∩B represents the overlapping web page set in the current user's web page set and the associated user's web page set;

[0024] tλ and tβ respectively represent the page weights corresponding to the pages in the overlapping page set in the historical operation set of the current user and the historical operation set of the associated user.

[0025] It means to obtain the overlapping page set in the current user's page set and the associated user's page set, and sum up the page weights corresponding to all the pages in the overlapping page set in the historical operation set of the current user and the page weights corresponding to them in the historical operation set of the associated user.

[0026] Among them, the first predetermined time period includes the predetermined time period before the current user executes the current search operation and / or the specific time period after the current user executes the current search operation;

[0027] The second predetermined time period is the predetermined time period before the current user executes the current search operation.

[0028] Among them, it further includes the following steps: after the current user executes the current search operation, every predetermined period, according to the user's web page browsing situation on the search page, update the page tags in step S2, and re-execute steps S3 to S7 according to the updated page tags.

[0029] Among them, after the current user executes the current search operation and before the page tag update operation is performed according to the user's web page browsing situation on the search page, determine the page tags through the keywords obtained in step S1.

[0030] Among them, it further includes the following steps: after the current user executes the current search operation, every predetermined period, according to the user's web page browsing situation on the search page, update the display ratios of the first display area and the second display area on the search page.

[0031] The present invention first roughly screens out recommended users for the current user through the browsing duration and browsing times of the web pages, and then, on the basis of the browsing duration and browsing times, further considers the user's operation situation on the web pages, and recommends the web pages that best meet the user's needs for the user through more refined screening indicators, greatly improving the accuracy of the recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is the logical thinking flow chart of the personalized recommendation preview method based on web page search of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] In order to have a further understanding of the technical solutions and beneficial effects of the present invention, the technical solutions of the present invention and the beneficial effects produced by them will be described in detail below with reference to the accompanying drawings.

[0034] The present invention aims at the following application scenarios to provide users with search convenience: In most search scenarios, users may have more than one search need. As the information collection work continues, users may need to continuously adjust keywords, etc., in order to obtain more comprehensive information. In daily search scenarios, taking the search for "tops" as an example, after the user browses enough links of tops, it is possible that the user needs to browse the shoes, pants, etc. that match them; taking the search for a certain movie as an example, after the user types in the movie name and browses enough movie reviews and news information related to the movie, it is possible that the user also needs to browse the corresponding theater information. In addition to daily search scenarios, when retrieving materials during study and work, as the query work progresses, after the user browses and studies some web pages, new problems may arise and new knowledge needs to be understood and learned. Thus, users need to continuously conduct associated searches. In most associated search scenarios, the associated web pages mostly belong to the same category. For example, the search web pages for tops, shoes, pants, etc. basically all belong to the "shopping" category; the search web pages for movie reviews, news, theaters, etc. basically all belong to the "entertainment" category; and when retrieving during study and work, the materials retrieved in one retrieval work basically all belong to the same discipline.

[0035] Therefore, the original intention of the present invention is that, in addition to providing a set of web pages for the search information typed in by the user, based on the user's historical operation information, other web pages related to the search information are recommended to the user, helping the user to diverge and expand the search idea, providing more comprehensive customized information, and continuously updating the search page based on the user's actual search operations.

[0036] Based on the above inventive concept, the search page of the present invention provides a first display area and a second display area; wherein, the first display area is used to display the web pages searched based on the search information typed in by the user; the second display area is used to display the web pages recommended to the user to help the user conduct associated searches based on the search information typed in by the user; the search methods of the web pages in the first display area and the second display area are described in detail below in conjunction with Figure 1 as shown and will be elaborated in the following text.

[0037] I. The First Display Area

[0038] In the present invention, a conventional web page search is performed based on the search information input by the current user, and the searched web pages are displayed in the first display area of the search page.

[0039] Specifically, the search information input by the current user may be a specific keyword or a search statement. For the search statement, it can be segmented through a word segmentation tool such as query to obtain keywords, and then a web page search is performed.

[0040] In the present invention, the webpage displayed in the first display area can be searched based on existing conventional search technology.

[0041] 2. Second display area

[0042] The web page displayed in the second display area is searched by the following method.

[0043] 1. Determine the web page tags

[0044] Web page tags represent the categories to which web pages belong, such as shopping, technology, news, air ticket booking, etc. Web page tags can be constructed in advance before the invention is implemented, and the vertical domains of web pages are matched with corresponding tags, and the corresponding relationship between the two is stored in a mapping table of the server.

[0045] In the present invention, after the user enters the search information, the web page tag can be determined based on the keywords obtained from the search information.

[0046] Specifically, the probability of each keyword belonging to each tag can be estimated based on the Bayesian classification method, the probability of each keyword belonging to each tag is summed up, and the tag with the largest probability value is determined as the webpage tag.

[0047] For example, if keyword 1, keyword 2, and keyword 3 are obtained for the search information typed by the user, and there are N preset web page tags, the probability sets of keyword 1 to keyword 3 belonging to the N web page tags are respectively:

[0048]

[0049] Sum the above three probability sets, and we get the following probability set:

[0050] ;

[0051] The label corresponding to the maximum numerical probability in the probability set can be determined as the web page label.

[0052] 2. Get the current user's web page collection

[0053] The current user web page set is a set of all web pages involving the determined web page tag browsed by the current user within the first predetermined time period.

[0054] Specifically, the first predetermined time period can be a predetermined time period before the execution of this search operation. The so-called "before the execution of this search operation" refers to a specific length of time period before the user types in the search information. For example, after the user types in "jeans", based on the web page tags obtained in step 1 above as "shopping", more specifically "shopping - clothing", all web pages related to "shopping - clothing" browsed by the user within a predetermined time (such as two months) before the execution of this search operation are obtained, and they are determined as the current user's web page set.

[0055] Specifically, through information such as the user's login name or identity identifier used to identify the user's identity, all web page sets browsed by the user within the first predetermined time period related to the determined web page tags are obtained. If the user has not browsed any web pages related to the determined web page tags within the first predetermined time period, the above-mentioned step 1 can be re-executed, and the web page tag corresponding to the second highest probability value is selected from the probability set as the determined web page tag, and so on.

[0056] 3. Determine associated users

[0057] Based on the current user's web page set, associated users are determined. Among them, an associated user is a user whose all web page sets browsed within the second predetermined time period related to the determined web page tags overlap with the current user's web page set. Correspondingly, all web page sets browsed by the associated user within the second predetermined time period related to the determined web page tags can be determined as the associated user's web page set. The number of associated users is two or more.

[0058] Specifically, the second predetermined time period can be a predetermined time period before the execution of this search operation, which is the same as above. The so-called "before the execution of this search operation" refers to a specific length of time period before the user types in the search information. The length of the first predetermined time period and the length of the second predetermined time period can be the same or different.

[0059] For example: Suppose the current user has browsed a certain web page 1 under the "shopping - clothing" category within two months before the execution of this search operation, and another user has also browsed a certain web page 1 under the "shopping - clothing" category within two months before the execution of this search operation by the current user. Then this user can be determined as an associated user of the current user.

[0060] During actual execution, some specific screening mechanisms can be set for associated users, including but not limited to:

[0061] (1) Determine the overlapping web page set between the associated user's web page set and the current user's web page set; the ratio of the number of web pages in the overlapping web page set to the number of web pages in the associated user's web page set should be greater than a predetermined value;

[0062] (2)Determine the overlapping web page set between the associated user's web page set and the current user's web page set; the ratio of the total browsing duration of the associated user on the web pages in the overlapping web page set to the total browsing duration of the associated user on the web pages in the associated user's web page set should be greater than a predetermined value;

[0063] (3)Determine the overlapping web page set between the associated user's web page set and the current user's web page set; the ratio of the number of web pages in the overlapping web page set to the total number of web pages in the associated user's web page set and the current user's web page set should be greater than a predetermined value.

[0064] In this way, it is ensured that the selected associated users have a certain degree of relevance, while reducing the data volume and the subsequent workload.

[0065] 4. Filter recommended users

[0066] Based on the historical operation set of the current user and the historical operation set of the associated user, determine the operation similarity between the current user and the associated user; based on the operation similarity, filter recommended users from the associated users.

[0067] The historical operation set of the current user includes each web page in the current user's web page set, the corresponding web page weight, and the operation vector of each web page; its storage form is (λ1, a1, ; λ2, a2, ; λ3, a3, ;...; λ n , a n , ), where a1, a2, a3... a n represent web pages, and λ1, λ2, λ3... λ n respectively represent the web page weights of web pages a1, a2, a3... a n , , , ... respectively represent the operation vectors of the current user on web pages a1, a2, a3... a n . Their meanings, determination methods, and uses will be described in detail below.

[0068] The historical operation set of the associated user includes each web page in the associated user's web page set, the corresponding web page weight, and the operation vector of each web page; the storage form is (β1, b1, ; β2, b2, ; β3, b3, ;...; β m , b m , ), where b1, b2, b3... b mDenote web pages, β1, β2, β3...β m respectively denote the web page weights of web pages b1, b2, b3...b m , , , ... respectively denote the operation vectors of the associated user for web pages b1, b2, b3...b m , and their meanings, determination methods, and uses will be described in detail below.

[0069] Among them, the web page weight is determined by the browsing duration and the number of browsing times. A relatively optimal determination method is as follows:

[0070] Q = Q inital + (x1 + T ime / C) / x2 + (y1 + C) / y2;

[0071] Among them, Q is the web page weight (representing λ1, λ2, λ3...λ n and β1, β2, β3...β m )), Q inital is the initial web page weight, x1, x2, y1, and y2 are adjustment factors, T ime is the total browsing duration of the corresponding user (the current user or the associated user) on the corresponding web page, and C is the total number of browsing times of the corresponding user (the current user or the associated user) on the corresponding web page.

[0072] By introducing the adjustment factors x1, x2, y1, and y2, a buffer is provided for the influence degrees of the average browsing duration and the total number of browsing times of the web page on the web page weight, avoiding the situation where the web page weight fluctuates greatly due to the changes in the average browsing duration and the total number of browsing times, and enhancing the stability of the recommendation preview method.

[0073] In the present invention, the operation similarity between the current user and the associated user is determined by the following formula:

[0074] ;

[0075] Among them, W represents the operation similarity between the current user and the associated user;

[0076] A represents the set of web pages of the current user, including web pages a1, a2, a3...a n ;

[0077] B represents the set of web pages of the associated user, including web pages b1, b2, b3...b m ;

[0078] n and m respectively represent the number of web pages in the set of web pages of the current user and the set of web pages of the associated user;

[0079] A ∩ B represents the set of web pages with overlap in the set of the current user's web pages and the set of associated users' web pages;

[0080] tλ and tβ respectively represent the web page weights corresponding to the web pages in the overlapping set of web pages in the historical operation set of the current user and the historical operation set of the associated user;

[0081] It means to obtain the set of overlapping web pages in the set of the current user's web pages and the set of associated users' web pages, and sum up the web page weights corresponding to all the web pages in the overlapping set of web pages in the historical operation set of the current user and the historical operation set of the associated user.

[0082] For example, assume that the historical operation set of the current user is (0.3, web page 1, ; 0.2, web page 2, ; 0.5, web page 3, ; 0.1, web page 4, ; 0.6, web page 5, ; 0.7, web page 6, ), and the historical operation set of a certain associated user is (0.5, web page 3, ; 0.2, web page 4, ; 0.6, web page 5, ; 0.8, web page 7, ; 0.5, web page 8, ; 0.9, web page 9, ); then in the historical operation set of the current user, the sum of the web page weights of all web pages = 0.3 + 0.2 + 0.5 + 0.1 + 0.6 + 0.7 = 2.4; in the historical operation set of this associated user, the sum of the web page weights of all web pages = 0.5 + 0.2 + 0.6 + 0.8 + 0.5 + 0.9 = 3.5; the sum of the web page weights of the overlapping web pages in the set of the current user's web pages and the set of associated users' web pages, that is, web page 3, web page 4, and web page 5 = (0.5 + 0.1 + 0.6) + (0.5 + 0.2 + 0.6) = 2.5; the operation similarity between the current user and this associated user = 2.5 / (2.4 + 3.5 - 2.5) = 0.74.

[0083] The meaning of the above formula is as follows: determine the overlapping web page set in the current user's web page set and the associated user's web page set, solve the sum of the web page weights in the overlapping web page set, and the ratio of the sum of the web page weights in the non-overlapping web page set in the current user's web page set and the associated user's web page set; since the web page weights integrate the browsing times and average browsing durations of relevant web pages, the ratio value obtained through the above calculation process can intuitively reflect the similarity degree of preferences of the current user and the associated user in browsing the web pages under the same web page label: the higher the ratio, the more inclined the current user and the associated user are to browse the same web pages when browsing the web pages under the same web page label, and the closer their preference degrees for different web pages are; based on this, subsequent web page recommendations for the associated user will be able to recommend web pages that better meet the needs of the current user.

[0084] Therefore, in the present invention, based on the historical operation set of the current user and the historical operation set of the associated user, the operation similarity between the current user and the associated user is determined; based on the operation similarity, the principle for screening recommended users from the associated users is: in the order from low to high of the operation similarity, successively eliminate the associated users with low operation similarity, and leave several associated users with the highest operation similarity for the subsequent generation of recommended web pages.

[0085] 5. Generate recommended web pages

[0086] Based on the historical operation set of the current user and the historical operation set of the recommended user, generate recommended web pages and display the recommended web pages in the second display area of the search page.

[0087] The storage form of the historical operation set of the current user has been described above. The recommended users are screened from the associated users. Therefore, the storage form of the historical operation set of the recommended users is the same as that of the historical operation set of the associated users above.

[0088] In the present invention, based on the operation situations of the current user and the recommended user on relevant web pages in the historical operation set, that is, the operation vectors, and combining the web page weights of the relevant web pages, recommended web pages are generated. A specific implementation method is as follows:

[0089] (1) Based on the operation vectors of each web page in the historical operation set of the current user, obtain the first absolute value of each web page in the historical operation set of the current user;

[0090] (2) Based on the operation vectors of each web page in the historical operation set of the recommended user, obtain the second absolute value of each web page in the historical operation set of the recommended user;

[0091] (3) Determine a set of recommended web pages, where the set of recommended web pages includes all web pages in the current user's web page set and the recommended user's web page set; the set of all web pages browsed by the recommended user within the second predetermined time period that involve the determined web page tags is the recommended user's web page set;

[0092] (4) Determine the recommended weight value of each web page in the set of recommended web pages through the following formula:

[0093] ;

[0094] where T is the recommended weight value of the web page;

[0095] λ i is the web page weight corresponding to the web page in the historical operation set of the current user, J1 is the first absolute value of the operation vector corresponding to the web page in the historical operation set of the current user. If there is no relevant web page in the historical operation set of the current user, then both λ i and J1 are set to 0;

[0096] β j is the web page weight corresponding to the web page in the historical operation set of the recommended user, J2 is the second absolute value of the operation vector corresponding to the web page in the historical operation set of the recommended user. If there is no relevant web page in the historical operation set of the recommended user, then both β j and J2 are set to 0.

[0097] (5) Screen out several web pages from the set of recommended web pages based on the recommended weight value as the final recommended web pages and display them in the second display area of the search page.

[0098] In the present invention, the operation vector is used to characterize what operations the user has performed on the web page during the process of browsing the web page, such as whether to like, comment, forward, collect, etc. The dimension of the operation vector is the number of types of the above operation types, and the values on each dimension can be represented by 1 or 0 to indicate whether the user has performed the relevant operation during the process of browsing the web page. For example: Suppose the operation vector has a total of four dimensions, respectively representing the operations of liking, commenting, forwarding, and collecting. If a user likes and comments on a certain web page but does not forward and collect it, then the operation vector of the user for this web page is (1, 1, 0, 0), and the absolute value of this operation vector is: .

[0099] For the above steps of generating recommended web pages, a specific example is given as follows: Suppose the historical operation set of the current user involves web pages 1, 2, 3, 4, and 5, and their respective web page weights are 0.1, 0.2, 0.3, 0.4, and 0.5. And the operations performed by the current user on each web page during the process of browsing web pages 1, 2, 3, 4, and 5 are as follows:

[0100] Webpage 1: Like, Comment, Forward, Favorite;

[0101] Webpage 2: Like, Comment;

[0102] Webpage 3: Like, Comment, Forward;

[0103] Webpage 4: Like, Comment, Favorite;

[0104] Webpage 5: Forward, Favorite;

[0105] Among the historical operation sets of a certain recommended user, webpages 3, 4, 6, 7, and 8 are involved, and their respective webpage weights are 0.9, 0.4, 0.6, 0.7, and 0.8. During the process of the recommended user browsing webpages 3, 4, 6, 7, and 8, the operations performed on each webpage are as follows:

[0106] Webpage 3: No operation;

[0107] Webpage 4: Like, Comment;

[0108] Webpage 6: Comment, Forward;

[0109] Webpage 7: Like, Comment, Forward;

[0110] Webpage 8: Forward, Favorite.

[0111] Then the storage forms of the historical operation set of the current user and the historical operation set of the recommended user are as follows:

[0112] (0.1, Webpage 1, (1, 1, 1, 1); 0.2, Webpage 2, (1, 1, 0, 0); 0.3, Webpage 3, (1, 1, 1, 0); 0.4, Webpage 4, (1, 1, 0, 1); 0.5, Webpage 5, (0, 0, 1, 1));

[0113] (0.9, Webpage 3, (0, 0, 0, 0); 0.4, Webpage 4, (1, 1, 0, 0); 0.6, Webpage 6, (0, 1, 1, 0); 0.7, Webpage 7, (1, 1, 1, 0); 0.8, Webpage 8, (0, 0, 1, 1)).

[0114] Then the recommended weights of each webpage in the recommended webpage set are as follows:

[0115]

[0116] After that, in the order of the recommended weight values from high to low, several webpages with high recommended weight values are selected from the recommended webpage set as the final recommended webpages and displayed in the second display area of the search page.

[0117] It should be noted that, taking Web page 3 as an example, since the recommended user did not perform any operations on this web page, it directly led to the value of "β j J2" being 0. However, as can be seen from the above text, in the historical operation set of this recommended user, the web page weight corresponding to Web page 3 is as high as 0.9, indicating that this recommended user browsed Web page 3 for a long time, or browsed it multiple times. Web page 3 should be a web page that better meets the needs of this recommended user and should also be a web page suitable for recommending to the current user. However, the second absolute value of Web page 3 is 0, resulting in the web page weight of this web page in the historical operation set of the recommended user not being considered at all when calculating the recommended weight value of this web page. In this way, it may lead to the omission of web pages that can provide important information.

[0118] Therefore, in a more preferred embodiment of the present invention, by introducing another index - the public absolute value, the recommended weight values of each web page are optimized.

[0119] Specifically as follows:

[0120] T = λ i J1 + β j J2; + (λ i + β j ) J / 2

[0121] Wherein, the meanings of T, λ i , β j , J1 and J2 are the same as those in the above text;

[0122] J is the public absolute value of the web page, and its determination method is to average the operation vectors of all historical operations of the web page before the current user performs this search operation (i.e., before typing the search information) to obtain the public operation vector, and the absolute value of the public operation vector is the public absolute value. For example: before the current user types the search information, a certain web page has been browsed 2000 times. Among the 2000 browsing records, it has been liked 2000 times, commented 1500 times, forwarded 1000 times, and collected 500 times. Then the public operation vector of this web page is (2000 / 2000, 1500 / 2000, 1000 / 2000, 500 / 2000) = (1, 0.75, 0.5, 0.25).

[0123] After screening out the recommended web pages, they should be checked for duplication with the web pages displayed in the first display area, and the recommended web pages that are the same as the web pages displayed in the first display area should be deleted. This is easily understood by those skilled in the art and will not be elaborated herein by the present invention.

[0124] So far, the web pages displayed in the first display area and the second display area have been screened out: Since the first display area conducts web page searches based on the user's search information, and the second display area conducts associated searches and recommendations based on web page tags, which is just an alternative provided to the user; at the very beginning of the search work, the web pages in the first display area should better meet the user's needs; therefore, the proportion of the initial first display area in the entire search page should be greater than the proportion of the second display area in the entire search page.

[0125] That is to say, the present invention first screens out recommended users with similar preferences to the current user from associated users. This screening process only considers the browsing duration and browsing times of the current user and associated users on web pages, which is equivalent to a rough screening process; then, based on the operation conditions of the current user and recommended users on web pages, recommended web pages are screened out. This screening process further considers operations such as liking, commenting, forwarding, and collecting on web pages by the current user and recommended users on the basis of browsing duration and browsing times, and at the same time introduces operations such as liking, commenting, forwarding, and collecting on web pages by the general public users. Through more refined screening indicators, web pages that best meet the user's needs are recommended for the user, greatly improving the accuracy of the recommendation.

[0126] III. Update

[0127] In the existing conventional search techniques, for the same keyword, the web pages given by existing search engines mostly involve many different classifications to provide users with more comprehensive web page information; however, since the second display area of the present invention recommends web pages based on web page tags, the recommended web pages only involve one web page tag; under the premise of inaccurate judgment of the classification to which the keyword belongs, it may lead to the situation that the recommended web pages are not what the user needs. For example, for the keyword "A Brief History of Time", it may represent the book "A Brief History of Time" or the movie "A Brief History of Time". When the user needs to search for the book "A Brief History of Time" and some physics knowledge related to this book, and the web pages displayed in the second display area are all related to the movie "A Brief History of Time", then the web pages recommended in the second display area have no value for the user at this time. Therefore, it is necessary to update the recommended web pages in the second display area based on the user's browsing situation on the search page.

[0128] A specific update method is as follows:

[0129] 1. After the current user performs this search operation, at regular intervals, update the web page tags according to the user's web page browsing situation on the search page.

[0130] In the present invention, the predetermined period can be a time period. For example, the web page tags are updated every five minutes. It can also be the browsing times period of the user for the web pages. For example, when the user browses the web pages 5 times, the web page tags are updated. Moreover, it can be the total browsing duration period of the user. For example, when the user's browsing time for the web pages reaches 20 minutes, the web page tags are updated.

[0131] The update principles of the web page tags can include the following methods:

[0132] (1) Obtain the web pages browsed by the current user within this predetermined period and the corresponding web page tags of each web page, count the number of web pages corresponding to each web page tag, and take the web page tag with the largest number of web pages as the updated web page tag;

[0133] (2) Obtain the web pages browsed by the current user within this predetermined period and the corresponding web page tags of each web page, count the total browsing duration of the current user on all the web pages corresponding to each web page tag, and take the web page tag with the largest total browsing duration as the updated web page tag.

[0134] 2. Based on the updated web page tags, sequentially re-determine the current user's web page set, associated users, and recommended users, and generate recommended web pages.

[0135] When determining the current user's web page set, it is necessary to obtain all the web page sets browsed by the current user within the first predetermined time period that involve the determined web page tags; when first obtaining the current user's web page set, as described above, the first predetermined time period is the predetermined time period before performing this search operation (at this time, the first predetermined time period is the same as the second predetermined time period); and when the user performs corresponding browsing operations on the web pages on the search page after performing this search operation, therefore, when re-determining the current user's web page set, the first predetermined time period can include the following situations:

[0136] (1) The first predetermined time period includes the predetermined time period before the current user performs this search operation and a specific time period after the current user performs this search operation; in this case, the historical browsing behavior of the current user before performing this search operation and the browsing behavior during this search operation will be considered;

[0137] (2) The first predetermined time period only includes a specific time period after the current user performs this search operation. In this case, only the browsing behavior of the current user during this search operation is considered.

[0138] However, when determining the associated users, it is necessary to obtain the users whose web page sets browsed within the second predetermined time period that involve the determined web page tags overlap with the current user's web page set; here, the second predetermined time period only includes the predetermined time period before the current user performs this search operation.

[0139] 3. Update the display ratios of the first display area and the second display area

[0140] As described above, the proportion of the initial first display area in the entire search page should be greater than that of the second display area in the entire search page.

[0141] However, as the user's search progresses, the web pages in the second display area are continuously updated based on the user's browsing behavior during the search operation, and the web pages in the second display area will become closer and closer to the user's needs. Therefore, similar to the rule for updating web page tags, every predetermined period, it is also necessary to update the display ratios of the first display area and the second display area on the search page according to the user's web page browsing situation on the search page.

[0142] The predetermined period for the update here can be the same as or different from the predetermined period for updating web page tags.

[0143] Similarly, similar to the update of web page tags, the predetermined period here can also be a time period, such as updating once every five minutes; it can also be the number of times the user browses the web page, such as updating once every 5 times the user browses the web page; it can even be the total browsing duration period of the user, such as updating once after the user's browsing time on the web page reaches 20 minutes.

[0144] The update principles for the display ratios of the first display area and the second display area on the search page may include the following methods:

[0145] (1) Obtain the number of web pages browsed by the current user in the first display area and the second display area during this predetermined period, and determine the display ratios of the first display area and the second display area on the search page based on the ratio of the number of web pages browsed by the current user in the first display area to the number of web pages browsed by the current user in the second display area;

[0146] (2) Obtain the duration of the web pages browsed by the current user in the first display area and the second display area during this predetermined period, and determine the display ratios of the first display area and the second display area on the search page based on the ratio of the total duration of the web pages browsed by the current user in the first display area to the total duration of the web pages browsed by the current user in the second display area.

[0147] Therefore, during the user's search process, the present invention continuously adjusts the display of the search page and recommends web pages based on the user's web page browsing situation, continuously adapts to the user's needs, and can provide a more user-friendly experience for the user.

[0148] Although the present invention has been described by using the above preferred embodiments, it is not intended to limit the protection scope of the present invention. Any person skilled in the art, without departing from the spirit and scope of the present invention, making various changes and modifications to the above embodiments still falls within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be defined by the claims.

Claims

1. A personalized recommendation preview method based on web search, characterized in that It includes the following steps: Step S1: Obtain keywords from the search information input by the current user, perform a web search based on the keywords, and display the searched web pages in the first display area of the search page; Step S2: Determine the web page tags, where the web page tags are the categories to which the web pages belong; Step S3: Obtain the current user's web page set, where the current user's web page set is the set of all web pages browsed by the current user within a first predetermined time period and related to the web page tags determined in Step S2; Step S4: Determine the associated users based on the current user's web page set; where the associated users are the users whose sets of all web pages browsed within a second predetermined time period and related to the web page tags determined in Step S2 overlap with the current user's web page set; Step S5: Determine the operation similarity between the current user and the associated users based on the historical operation set of the current user and the historical operation set of the associated users; Step S6: Screen and recommend users from the associated users based on the operation similarity; Step S7: Generate recommended web pages based on the historical operation set of the current user and the historical operation set of the recommended users, and display the recommended web pages in the second display area of the search page; Among them, the recommended web pages are generated through the following steps: Step S71: Obtain the first absolute values of the web pages in the historical operation set of the current user based on the operation vectors of the web pages in the historical operation set of the current user; Step S72: Obtain the second absolute values of the web pages in the historical operation set of the recommended users based on the operation vectors of the web pages in the historical operation set of the recommended users; Step S73: Determine a set of recommended web pages, where the set of recommended web pages includes all the web pages in the current user's web page set and the recommended user's web page set; the set of all web pages browsed by the recommended user within a second predetermined time period and related to the web page tags determined in Step S2 is the recommended user's web page set; Step S74: Determine the recommended weight values of the web pages in the set of recommended web pages through the following formula: T = λ i J1 + β j J2 + (λ i + β j ) J / 2 λ i is the web page weight corresponding to the web page in the historical operation set of the current user. J1 is the first absolute value of the operation vector corresponding to the web page in the historical operation set of the current user. If there is no relevant web page in the historical operation set of the current user, then both λ i and J1 are taken as 0; β j is the web page weight corresponding to the recommended user's historical operation set, and J2 is the second absolute value of the operation vector corresponding to the web page in the recommended user's historical operation set. If there is no relevant web page in the recommended user's historical operation set, then β j and J2 are both set to 0; J is the absolute value of the public for the web page; Step S75: Screen out several web pages from the set of recommended web pages based on the recommended weight values as the final recommended web pages.

2. The personalized recommendation preview method based on web search according to claim 1, wherein The historical operation set of the current user includes each web page in the current user's web page set and the web page weight corresponding to each web page; The historical operation set of the associated users includes each web page in the associated user's web page set and the web page weight corresponding to each web page; The historical operation set of the recommended users includes each web page in the recommended user's web page set and the web page weight corresponding to each web page; The associated user's web page set is the set of all web pages browsed by the associated user within a second predetermined time period and related to the web page tags determined in Step S2; The web page weight is determined by the browsing duration and the number of browsing times.

3. The personalized recommendation preview method based on web search according to claim 2, characterized in that The operation similarity between the current user and the associated users is determined through the following formula: W= ； Where W represents the operation similarity between the current user and the associated users; A represents the set of current user web pages, including web pages a1, a2, a3... a n , web pages a1, a2, a3... a n The page weights of a1, a2, a3... a are λ1, λ2, λ3... λ n ; B represents the set of associated user web pages, including web pages b1, b2, b3... b m , web pages b1, b2, b3... b m The page weights of are β1, β2, β3... β m ; n and m respectively represent the number of web pages in the current user's web page set and the associated user's web page set; A∩B represents the set of overlapping web pages in the current user's web page set and the associated user's web page set; $t\lambda$ and $t\beta$ respectively represent the page weights corresponding to the pages in the overlapping page set in the historical operation set of the current user and the historical operation set of the associated user. It means to obtain the overlapping web page set between the current user's web page set and the associated user's web page set, and sum up the web page weights corresponding to all web pages in the overlapping web page set in the current user's historical operation set and the web page weights corresponding to them in the associated user's historical operation set.

4. The personalized recommendation preview method based on web search according to claim 1, characterized in that: The first predetermined time period includes a predetermined time period before the current user executes the current search operation and / or a specific time period after the current user executes the current search operation. The second predetermined time period is a predetermined time period before the current user executes the current search operation.

5. The personalized recommendation preview method based on web search according to claim 4, wherein It further includes the following steps: After the current user executes the current search operation, every predetermined cycle, according to the user's web page browsing situation on the search page, update the web page tags in step S2, and re-execute steps S3 to S7 according to the updated web page tags.

6. The personalized recommendation preview method based on web search according to claim 5, wherein: After the current user executes the current search operation and before the web page tag update operation is performed according to the user's web page browsing situation on the search page, determine the web page tags through the keywords obtained in step S1.

7. The personalized recommendation preview method based on web search according to claim 4, characterized in that It further includes the following steps: After the current user executes the current search operation, every predetermined cycle, according to the user's web page browsing situation on the search page, update the display ratios of the first display area and the second display area on the search page.

Images