CN102316167B - Website recommending method, system thereof and network server - Google Patents

Abstract

The present invention provides a website recommendation method and system and a network server, wherein the method includes: the network server obtains the characteristic information of the website visited by the user according to the user's Internet access information within a preset time period, and clusters and acquires the users according to the characteristic information Multiple user clusters, so that when receiving a network access request including a user ID sent by a user terminal, it is judged whether the user includes the first user corresponding to the user ID, and if so, according to the other users in the user cluster where the first user is located The feature information determines the website recommended to the first user, and embeds the website address of the recommended website into the network access response and returns it to the user terminal, realizing that the network server can recommend to the user who conducts network access based on the global user network access behaviorMore website, so that users can obtain more interesting information.

Description

Website recommendation method and system and web server

technical field

The invention relates to communication technology, in particular to a website recommendation method and system and a network server.

Background technique

With the development of electronic information technology, the network has changed people's way of life. For example, people can use the network to obtain books, movies, music, and even commodities that they are interested in. Therefore, the network has brought us an efficient and convenient life. , People have become accustomed to using computers, mobile phones and other devices with Internet access capabilities to browse the web pages they are interested in for learning, entertainment, and shopping to meet their own multi-faceted needs.

People can use the Internet to obtain rich information more efficiently for learning and entertainment. Specifically, the web server will recommend related websites of the same type to the user according to the type of website visited by the user for the user's reference. For the type of website, the web server will recommend other websites in the information technology category to the user for reference, and the web server will record the websites frequently visited by the user and obtain relevant website recommendations to the user, so that the user can obtain more interesting information .

However, the network server in the prior art only recommends relevant websites to the user for reference according to the user's own network access behavior, so that the information obtained by the user is limited and has certain limitations.

Contents of the invention

Aiming at the above-mentioned defects in the prior art, embodiments of the present invention provide a website recommendation method and system and a web server.

An embodiment of the present invention provides a website recommendation method, including:

The network server obtains the characteristic information of the website visited by the user who accesses the network within the preset time period according to the locally stored Internet access information;

The network server performs clustering analysis on the users according to the feature information to obtain multiple user clusters, so as to determine whether the user includes a user ID related to the user ID when receiving a network access request sent by the user terminal. For the corresponding first user, if yes, determine the website recommended to the first user according to the feature information of the users in the user cluster where the first user is located, and embed the website address of the recommended website into the network access The response is returned to the user terminal.

An embodiment of the present invention provides a network server, including:

The first obtaining module is used to obtain the characteristic information of the website visited by the user who visits the network within a preset time period according to the locally stored online information;

The second acquisition module is configured to perform cluster analysis on the users according to the feature information to acquire multiple user clusters;

A judging module, configured to judge whether the user includes a first user corresponding to the user identifier when receiving a network access request including the user identifier sent by the user terminal;

A processing module, configured to determine, according to the characteristic information of the users in the user cluster where the first user is located, which user is recommended to the first user, if it is determined that the user includes the first user corresponding to the user identifier. website, and embed the URL of the recommended website into the network access response and return it to the user terminal.

An embodiment of the present invention provides a website recommendation system, including the above-mentioned network server and a user terminal.

In the website recommendation method and system and the web server provided by the embodiments of the present invention, the web server obtains the characteristic information corresponding to the website visited by the user according to the user's Internet access information within a preset time period, and clusters the users according to the characteristic information to obtain multiple users. cluster, so that when receiving a network access request including a user ID sent by a user terminal, if it is determined that the user includes the first user corresponding to the user ID, then according to the feature information of the remaining users in the user cluster where the first user is located, determine the direction to The website recommended by the first user, and the website address of the recommended website is embedded in the network access response and returned to the user terminal, so that the network server can recommend more websites to the user who conducts network access based on the overall network access behavior of the user, thereby To enable users to obtain more interesting information.

Description of drawings

Fig. 1 is the flowchart of an embodiment of website recommendation method of the present invention;

Fig. 2 is a flowchart of another embodiment of the website recommendation method of the present invention;

Fig. 3 is a schematic structural diagram of an embodiment of the network server of the present invention;

FIG. 4 is a schematic structural diagram of an embodiment of the website recommendation system of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Fig. 1 is the flow chart of an embodiment of website recommendation method of the present invention, as shown in Fig. 1, this method comprises:

Step 100, the network server obtains the feature information of the website visited by the user who visits the network within a preset time period according to the locally stored online information;

Users can send network access requests to the network server through mobile phones, computers and other user terminals with Internet access functions. It can be understood that, in this embodiment, the refresh time of the network server is set according to specific application needs, such as three days or one week. The user's online information stored by the web server specifically includes: the user's user ID, the website visited each time, and the corresponding start time and end time.

It should be noted that the user identification in this embodiment is used to distinguish different users. Those skilled in the art can understand that there are many forms of user identification according to different application scenarios and different information processing methods. This example does not limit the specific form of user identification. For example, in a local area network, the user terminal used by each user has a fixed IP address to identify different users, and the user identification in the user's online information stored on the network server of the local area network is the IP address of the user terminal; or the local area network In order to ensure the security of users’ network access, users are required to access the network only after identity information authentication through an external electronic device. Therefore, the user ID in the user’s online information stored on the network server of the local area network Can be the user's identity information.

The network server obtains the feature information corresponding to the website accessed by the user who accesses the network within the preset time period according to the locally stored Internet access information. It should be noted that the feature information in this embodiment reflects The behavior characteristics of visited users visiting websites may specifically include the frequency characteristics of each user visiting each website within a preset time period and/or the time period characteristics of each visit. Wherein, the preset time period in this embodiment can be set in advance on the network server according to specific application conditions. For example, if the preset time period is from 0:00 to 24:00 every day, that is, the network server will The stored surfing information is statistically obtained to obtain the frequency characteristics of each website visited by each user who has accessed the network within the previous 24 hours and/or the time period characteristics of each visit. In the specific implementation process, the network server will perform analog-to-digital conversion on the obtained frequency feature or time period feature, and if the feature information includes the frequency feature and time period feature, weight the digital quantities of the two to obtain the corresponding feature information. It is worth noting that the feature information in this embodiment is not limited to the frequency feature and time period feature, and can also be adjusted according to the obtained specific online information to obtain other feature information. The specific processing process is as above, and will not be repeated here. . In order to illustrate the meaning of feature information more clearly, an example is shown in Table 1. Table 1 shows the feature information corresponding to the website visited by users who access the network within a preset time period. The feature information is for each user in the preset time period. Set the value obtained after the weighted modulus conversion of the frequency characteristics and time period characteristics of visiting each website within the time period.

Table 1

Step 101, the network server clusters the users according to the feature information to obtain multiple user clusters;

The network server performs cluster analysis on all users who access the network on the network server within the time period according to the obtained feature information corresponding to the websites accessed by the users who access the network within the preset time period to obtain multiple user clusters. Cluster Analysis, also known as group analysis, is a process of classifying data into different classes or clusters, so objects in the same cluster have great similarity, while objects in different clusters have great similarity. dissimilarity. The calculation methods of cluster analysis mainly include partitioning methods, hierarchical methods, density-based methods, grid-based methods and model-based methods. -based methods). The specific implementation process of each clustering method belongs to the existing technology. In order to explain the process of clustering analysis more clearly, the K-means in the splitting method is used as an example to explain in detail, and the rest of the clustering methods will not be repeated one by one. .

The algorithm of K-means introduced in Table 1 above is as follows:

Step (1): When the user cluster k=2 is taken as an example, two users are randomly selected from users 1 to 4 as initial centroids (centers of categories), assuming that users 1 and 2 are selected;

Step (2): For each remaining user, calculate its distance to each centroid: suppose the calculation result from user 3 to user 1 is a; suppose the calculation result from user 3 to user 2 is b, if a<b, Then the distance between user 3 and user 1 is closer, and user 3 and user 1 are divided into one class; and so on for the remaining users, and finally all users can be divided into two groups with user 1 and user 2 as the centroid. class;

Step (3): Recalculate the centroid for each class. The calculation method is to average the weights of each user. After calculating the new centroid of each class, for all users, calculate the distance to each centroid, and so on. , until the center of mass no longer changes.

Step (4): For each class, calculate the mean square error within the class, that is, the distance from all users in the class to the centroid, and compare their mean square errors. The trend should be gradually decreasing. When the mean square error value drops significantly to no Then the significantly decreased K value can be used as the final K, that is, the number of user clusters.

Step 102, when receiving a network access request including a user ID sent by a user terminal, determine whether the user includes a first user corresponding to the user ID, and if so, then according to the user cluster in which the first user is located The feature information of the remaining users determines the recommended website to the first user, and embeds the website address of the recommended website into a network access response and returns it to the user terminal.

When the network server receives the network access request including the user ID sent by the user terminal, it queries the clustered users according to the user ID to determine whether the first user corresponding to the user ID is included. If it is determined that the first user is included in the clustered users, it means that the first user has also undergone cluster analysis, and the user cluster obtained in step 101 is queried according to the user ID and the user cluster where the first user is located is determined. Based on the above It can be learned that the users in the user cluster have similar network access behaviors to the first user. Obtain the feature information of the users in the user cluster where the first user is located, and determine the website recommended to the first user according to the set recommendation rules according to the feature information. Websites visited within a preset time period, and websites visited by other users except the first user but not visited by the first user are recommended to the first user. It should be noted that the recommendation rules are specifically set according to specific application scenarios, and this embodiment does not limit the specific recommendation rules.

The web server embeds the web address of the website recommended to the first user into the web access response and returns it to the user terminal. Wherein, the URL of the website includes domain name and/or IP address, and the URL of the website visited by the user in the online information stored on the network server is represented by the domain name or IP address. The URL is an IP address, and the web server can directly embed the IP address into the network access response and return it to the user terminal, or send a domain name reverse query request including the IP address to the domain name server, and the domain name server sends a query to the web server through PTR domain name resolution. The domain name corresponding to the IP address is returned, and the network server embeds the IP address and the corresponding domain name of the website into the network access response and returns it to the user terminal for the first user's reference, and returns the domain name to the user terminal, which is convenient for the user to remember and write, thereby Make it easier for users to search and visit recommended websites. If it is determined based on Internet access information that the website address recommended to the first user is a domain name, the network server may directly embed the domain name into the network access response and return it to the user terminal, or may send a domain name query request including the domain name to the domain name server, and the domain name server The IP address corresponding to the domain name is returned to the network server through A-type domain name resolution, and the network server embeds the IP address of the website and the corresponding domain name into the network access response and returns it to the user terminal for reference by the first user, and returns to the user terminal IP address, so that users can search and access recommended websites more directly, without initiating a domain name query request to the domain name server.

In the website recommendation method provided in this embodiment, the network server obtains the feature information corresponding to the website visited by the user who accesses the network within a preset time period according to the user's Internet access information, and performs cluster analysis on the user according to the feature information to obtain multiple users. cluster, when the network server receives a network access request including a user ID, if it is judged that the clustered users include the first user corresponding to the user ID, then according to the feature information of the user in the user cluster where the first user is located, determine the direction to The website recommended by the first user, and the website address of the recommended website is embedded in the network access response and returned to the user terminal, so that the network server can recommend more websites to the user who conducts network access based on the overall network access behavior of the user, thereby To enable users to obtain more interesting information.

Fig. 2 is a flowchart of another embodiment of the website recommendation method of the present invention, as shown in Fig. 2, the method includes:

Step 200, the network server obtains the feature information of the website visited by the user who visits the network within a preset time period according to the locally stored online information;

Step 201, the network server clusters the users according to the feature information to obtain multiple user clusters;

Step 202, when the network server receives the network access request including the user ID sent by the user terminal, it judges whether the user includes the first user corresponding to the user ID, and if not, broadcasts the request including the user ID to other network servers. The user identifier and the online information query request of the preset time period, if the online information of the first user returned by the other network servers within the preset time period is received, the online information is obtained according to the online information. Describe the feature information corresponding to the website visited by the first user;

When the network server receives the network access request including the user ID sent by the user terminal, it queries the clustered users according to the user ID to determine whether the first user corresponding to the user ID is included. If it is determined that the clustered users do not include the first user, it means that the first user has not accessed the network through the network server within the preset time period, that is to say, the network server has not accessed the network within the preset time period. The users performing network access in excluding the first user.

The network server broadcasts to other network servers in the Internet system an online information query request including the user ID of the first user and a preset time period. If the network server can receive the first user's Internet access information returned by other network servers within the preset time period from the locally stored online information within the preset time period information, according to the online information of the first user, the characteristic information corresponding to the website visited by the first user is obtained. For the specific characteristic information acquisition process, refer to step 100 in the first embodiment above, which will not be repeated here.

Step 203, the network server obtains the corresponding aggregated profile information according to the feature information of the users in each user cluster, and determines the corresponding aggregated profile information according to the feature information and the aggregated profile information of the first user through a similarity measure. The user cluster to which the first user belongs;

The web server obtains corresponding aggregated profile information according to the feature information of users in each user cluster obtained in the above step 201, where the aggregated profile information is the average weight of the feature information corresponding to the website visited by the user in each user cluster;

The network server performs similarity measurement according to the feature information of the first user and the acquired aggregated profile information. It should be noted that there are many methods for similarity measurement, such as Pearson correlation coefficient or cosine coefficient, which are not specifically limited in this embodiment. The matching score between the feature information corresponding to the website visited by the first user and each aggregated profile information is obtained through the similarity measure to determine the user cluster to which the first user belongs. The larger the matching score, the better the relationship between the first user and the users in the user cluster. The higher the similarity, the user cluster with the largest matching score is selected as the user cluster to which the first user belongs.

Step 204: Determine the website recommended to the first user according to the characteristic information of the users in the user cluster where the first user belongs, and embed the website address of the recommended website into the network access response and return to the the user terminal.

Obtain the feature information of the users in the user cluster where the first user is located, and determine the website recommended to the first user according to the set recommendation rules according to the feature information, specifically, the remaining users in the user cluster where the first user is located The feature information is weighted and averaged to obtain the recommendation scores of the websites visited by other users. According to the recommendation scores of each website visited by other users, the website recommended to the first user is determined according to the preset recommendation criteria. For example, according to the recommendation score of each website from The websites are arranged from high to low and these websites have not been visited by the first user, until the number of preset recommended websites is reached, and the selected websites are used as the websites recommended to the first user. The web server embeds the web address of the recommended website into the web access response and returns it to the user terminal for reference by the first user. Refer to the above-mentioned embodiment for the specific process, which will not be repeated here.

For the specific implementation process of step 201 and step 202 in this embodiment, refer to the embodiment shown in FIG. 1 , which will not be repeated here.

In the website recommendation method provided in this embodiment, the network server obtains the feature information corresponding to the website visited by the user who accesses the network within a preset time period according to the user's Internet access information, and performs cluster analysis on the user according to the feature information to obtain multiple users. Clustering, when the network server receives a network access request that includes a user ID, if it is judged that the clustered users do not include the first user corresponding to the user ID, it broadcasts queries to other network servers, and if the other network servers receive The returned online information of the first user determines the user cluster where the first user is located, and determines the website recommended to the first user according to the feature information of the user in the user cluster where the first user is located, and embeds the website address of the recommended website The network access response is returned to the user terminal, so that the network server can further recommend more websites to the network accessing user based on the overall user network access behavior, so that the user can obtain more interesting information.

Those of ordinary skill in the art can understand that all or part of the steps for realizing the above-mentioned method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the It includes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

Fig. 3 is a schematic structural diagram of an embodiment of the network server of the present invention. As shown in Fig. 3, the network server includes: a first acquisition module 11, a second acquisition module 12, a judging module 13 and a processing module 14, wherein the first acquisition Module 11 is used to obtain the characteristic information of the user visiting the website that performs network access within a preset period of time according to the online information stored locally; the second acquisition module 12 is used to perform cluster analysis on users according to the characteristic information to obtain multiple user clusters; judge Module 13 is used to determine whether the user includes the first user corresponding to the user ID when receiving a network access request sent by the user terminal that includes the user ID; the processing module 14 is used to determine whether the user includes the first user corresponding to the user ID , the website recommended to the first user is determined according to the feature information of the users in the user cluster where the first user is located, and the website address of the recommended website is embedded into the network access response and returned to the user terminal.

For the embodiment shown in FIG. 3 , the second acquisition module 12 can perform cluster analysis on users by splitting method, hierarchical method, density-based method, grid-based method and model-based method according to feature information.

For the functions and processing flow of each module in the network server provided in this embodiment, refer to the method embodiment shown in FIG. 1 above. The implementation principles and technical effects are similar, and will not be repeated here.

Based on the embodiment shown in FIG. 3 , further, the processing module 14 is further configured to, if it is judged that the learned user does not include the first user corresponding to the user ID, then broadcast the Internet access information including the user ID and the preset time period to other network servers Inquiry request, if the first user’s online information returned by other network servers within the preset time period is received, obtain the characteristic information corresponding to the website visited by the first user according to the online information; obtain according to the characteristic information of the users in each user cluster corresponding aggregate profile information, and determine the user cluster to which the first user belongs according to the feature information of the first user and the aggregate profile information through a similarity measure.

For the functions and processing flow of each module in the network server provided in this embodiment, refer to the method embodiment shown in FIG. 2 above. The implementation principles and technical effects are similar, and will not be repeated here.

Fig. 4 is a schematic structural diagram of an embodiment of the website recommendation system of the present invention. As shown in Fig. 4, the system includes: a web server 1 and a user terminal 2, wherein the web server 1 may be the web server provided by the embodiment of the present invention, and the user Terminal 2 is the user terminal involved in the embodiment of the present invention. For the functions and processing flow of each device in the website recommendation system provided by this embodiment, refer to the above-mentioned method and device embodiments. The implementation principles and technical effects are similar, and are not described here. Let me repeat.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (8)

1. a recommendation of websites method, is characterized in that, comprising:

The webserver obtains the characteristic information of user's access websites of carrying out access to netwoks in the Preset Time section according to the internet information of this locality storage;

The described webserver carries out cluster analysis according to described characteristic information to described user and obtains a plurality of users bunch, so that receive that user terminal sends comprise the network access request of user ID the time, judge whether described user comprises the first user corresponding with described user ID, if, determine the website of recommending to described first user according to the described characteristic information of the user in the user at described first user place bunch, and the network address of the website of described recommendation is embedded in the access to netwoks response returns to described user terminal;

If judgement knows that described user does not comprise the first user corresponding with described user ID, described method also comprises:

The described webserver comprises the internet information query requests of described user ID and described Preset Time section to all the other webserver broadcasting, if receive the internet information of described first user in described Preset Time section that described all the other webservers return, obtain the website characteristic of correspondence information of described first user access according to described internet information;

The described webserver obtains corresponding gathering profile information according to the described characteristic information of user in each user bunch, and determines user bunch under described first user according to the described characteristic information of described first user and described gathering profile information by similarity measurement.

2. recommendation of websites method according to claim 1, is characterized in that, describedly according to described characteristic information, described user carried out cluster analysis and comprise:

According to described characteristic information by disintegrating method, stratification, density-based method, based on method and the model-based methods of grid, described user is carried out cluster analysis.

3. recommendation of websites method according to claim 1, is characterized in that, the described characteristic information of all the other users in described user according to described first user place bunch is determined to comprise to the website that described first user is recommended:

Obtain the recommender score of the website of described all the other users' access according to the described characteristic information of all the other users in the user at described first user place bunch;

Determine to the website of described first user recommendation according to default recommendation criterion according to described recommender score.

4. according to claim 1-3 described recommendation of websites methods of any one, is characterized in that, described characteristic information comprises: the characteristics of time interval of the frequecy characteristic of described user's access websites and/or each access in the Preset Time section.

5. according to claim 1-3 described recommendation of websites methods of any one, is characterized in that, described network address comprises: domain name and/or IP address.

6. a webserver, is characterized in that, comprising:

The first acquisition module is used for obtaining according to the internet information of this locality storage the characteristic information of user's access websites of carrying out access to netwoks in the Preset Time section;

The second acquisition module is used for according to described characteristic information, described user being carried out cluster analysis and obtains a plurality of users bunch;

Judge module, be used for receive that user terminal sends comprise the network access request of user ID the time, judge whether described user comprises the first user corresponding with described user ID;

Processing module, know that described user comprises the first user corresponding with described user ID if be used for judgement, determine the website of recommending to described first user according to the described characteristic information of user in the user at described first user place bunch, and the network address of the website of described recommendation is embedded in the access to netwoks response returns to described user terminal;

Described processing module also is used for:

If judgement knows that described user does not comprise the first user corresponding with described user ID, comprise the internet information query requests of described user ID and described Preset Time section to all the other webserver broadcasting, if receive the internet information of described first user in described Preset Time section that described all the other webservers return, obtain the website characteristic of correspondence information of described first user access according to described internet information;

Obtain corresponding gathering profile information according to the described characteristic information of user in each user bunch, and determine user bunch under described first user according to the described characteristic information of described first user and described gathering profile information by similarity measurement.

7. the webserver according to claim 6, is characterized in that, described the second acquisition module specifically is used for:

According to described characteristic information by disintegrating method, stratification, density-based method, based on method and the model-based methods of grid, described user is carried out cluster analysis.

8. a recommendation of websites system, is characterized in that, comprises the webserver as described in claim 6 or 7, and user terminal.

Images