CN104462438A

CN104462438A - Information processing method and device

Info

Publication number: CN104462438A
Application number: CN201410778890.9A
Authority: CN
Inventors: 梁爽; 廉磊
Original assignee: Beijing Sogou Technology Development Co Ltd
Current assignee: Beijing Sogou Technology Development Co Ltd
Priority date: 2014-12-15
Filing date: 2014-12-15
Publication date: 2015-03-25

Abstract

The invention relates to the field of Internet, discloses an information processing method and device and aims at solving the technical problem that equipment needs to acquire a lot of useless data from a network in the prior art. The information processing method includes when an access operation accessing a first webpage is detected, judging whether the first webpage is a webpage of a preset type or not; when the first webpage is the webpage of the preset type, acquiring description information of a target object corresponding to the first webpage; determining first recommending information related to the target object through the description information, and outputting the first recommending information. The information processing method and device has the advantage of lowering volume of data that the equipment acquires from the network.

Description

A kind of information processing method and device

Technical field

The present invention relates to internet arena, particularly relate to a kind of information processing method and device.

Background technology

Nowadays, the Internet era that people entering, internet provides a large amount of information, user can be searched for keyword on the internet by search engine, obtains the information needed for user.

And along with the development of Internet technology, user can obtain increasing data from web search, books reading website is used for user, user is when checking certain this e-book, first the evaluation information of this e-book can be checked, then from the evaluation of web search for this e-book, and then whether this e-book is worth reading; And do shopping at shopping at network platform for user; usual user can browse several shops usually by shopping at network platform; for a certain network shop; whether the information such as merchandise news, evaluation information, service, logistics, anonymous ratings that user understands browse network shop is usually reliable to determine this family's network shop; then also can browse buyer's guide information, evaluation information for a certain commodity, and sometimes also can search for some merchandise related information and then determine whether to buy this commodity on network.

From analyzing above, user is when obtaining corresponding informance by certain destination object (such as: e-book, commodity etc.) search, need access a lot of pages, but a lot of data of the page in the page of accessing the data required for non-user, and user also needs to carry out a large amount of induction-arrangements, as can be seen here, when also existing in prior art from network acquisition information, equipment needs the technical matters from a large amount of gibberish of Network Capture, and then it is excessive to cause data to transmit burden, and consuming time long.

Summary of the invention

The embodiment of the present application is by providing a kind of information processing method and device, and the equipment in prior art that solves needs the technical matters from a large amount of gibberish of Network Capture.

First aspect, the embodiment of the present invention provides a kind of information processing method, comprising:

When the accessing operation of access first webpage being detected, judge that whether described first webpage is the webpage of preset kind;

When described first webpage is the webpage of described preset kind, obtain the descriptor of the corresponding destination object of described first webpage;

Determined to there is the first recommendation information associated with described destination object by described descriptor, and export described first recommendation information.

Optionally, the webpage of described preset kind is specially: shopping class webpage; The descriptor of the corresponding destination object of described first webpage of described acquisition, is specially:

Obtain the descriptor of at least one commodity of the network shop corresponding to described first webpage, at least one commodity described are described destination object;

Describedly determined to there is the first recommendation information associated with described destination object by described descriptor, be specially:

Determine to there is with the commodity of described network shop or described network shop described first recommendation information associated by described descriptor.

Optionally, described descriptor is specially: the buyer's guide information of at least one commodity described; Or user is to the evaluation information of at least one commodity described.

Optionally, when described first page is the sales page of the first commodity, described descriptor is specially: the buyer's guide information of described first commodity or user are to the evaluation information of described first commodity.

Optionally, be: during the buyer's guide information of at least one commodity described describedly determine to there is with the commodity of described network shop or described network shop the first recommendation information associated by described descriptor in described descriptor, specifically comprise:

The shop merchandise classification information of the commodity that described network shop is sold is determined by the buyer's guide information of at least one commodity described;

Carry out search using described shop merchandise classification information as keyword and determine described first recommendation information, the recommendation information that described first recommendation information is corresponding with first network link information.

Optionally, described first recommendation information is specially: the keyword of described first network link information and described first network link information; Or the content of pages corresponding to described first network link information; Or the information obtained after the content of pages of described first network link information summarized.

Optionally, describedly carry out search using described shop merchandise classification information as keyword and determine described first recommendation information, be specially:

From the first corresponding relation prestored, obtain described first recommendation information by described shop merchandise classification information searching, in described first corresponding relation, include the corresponding relation of keyword and recommendation information; Or

Carry out search using described shop merchandise classification information as keyword in a network and obtain described first network link information, and determine described first recommendation information by described first network link information.

Optionally, be specially in described descriptor: during the evaluation information of described first commodity, describedly determined to there is the first recommendation information associated with the commodity of described network shop by described descriptor, specifically comprise:

Analysis is carried out to the evaluation information of described first commodity and determines the first keyword set;

The keyword of the first kind is determined from described first keyword set, the keyword of the described first kind is specially: carry out the keyword of unfavorable ratings to described first commodity or described first commodity carried out to the keyword of front evaluation, the keyword of the described first kind is described first recommendation information.

Optionally, the described keyword determining the first kind from described first keyword set, specifically comprises:

Determine the type of merchandise information of described first commodity;

Based on the type of merchandise information of described first commodity, from the keywords database of multiple type of merchandise, determine the first keywords database;

Judge in described first keyword set, whether each keyword is positioned at described first keywords database one by one;

When a certain keyword in the first keyword set is positioned at described first keywords database, then determine that it is the keyword of the described first kind.

Optionally, be specially in described descriptor: during the buyer's guide information of described first commodity, describedly determined to there is the first recommendation information associated with the commodity of described network shop by described descriptor, specifically comprise:

The type of merchandise information of described first commodity and the brand message of described first commodity is determined by the buyer's guide information of described first commodity;

Using the brand message of the type of merchandise information of described first commodity and described first commodity as keyword, search obtains described first recommendation information, and described first recommendation information is: the recommendation information corresponding with second network link information.

Optionally, described first recommendation information is specially: the keyword of described second network link information and described second network link information; Or the content of pages corresponding to described second network link information; Or the information obtained after the content of pages of described second network link information summarized.

Optionally, the described brand message using the type of merchandise information of described first commodity and described first commodity obtains described first recommendation information as keyword, specifically comprises:

From the second corresponding relation prestored, obtain described first recommendation information by the type of merchandise information of described first commodity and the brand message of described first commodity as keyword lookup, in described second corresponding relation, include the corresponding relation of keyword and recommendation information; Or

Carry out the described second network link information of search acquisition using the brand message of the type of merchandise information of described first commodity and described first commodity in a network as keyword, and determine described first recommendation information by described second network link information.

Optionally, describedly determine to there is with the commodity of described network shop or described network shop described first recommendation information associated by described descriptor, specifically comprise:

The second keyword set is obtained out from described descriptor;

Judge whether a certain keyword in described second keyword set is arranged in a certain predetermined keyword list of multiple predetermined keyword list one by one;

When the first keyword in described second keyword set is positioned at the first predetermined keyword list, determine described first recommendation information comprising fisrt feature information, described fisrt feature information is corresponding with described first predetermined keyword list.

Optionally, when described the first keyword in described second keyword set is positioned at the first predetermined keyword list, determines described first recommendation information comprising fisrt feature information, specifically comprise:

When described first keyword is positioned at described first predetermined keyword list, determine the number percent of keyword shared by described second keyword set in described first predetermined keyword list;

Judge whether described number percent is greater than predetermined threshold value;

When described number percent is greater than described predetermined threshold value, described fisrt feature information is added described first recommendation information.

Optionally, described method also comprises:

When described first webpage is described shopping class webpage, obtains and export the second recommendation information, described second recommendation information is specially: the negative Transaction Information of described network shop.

Optionally, described negative Transaction Information specifically comprises: the commodity sold of described network shop by the quantity of reimbursement, described network shop the complained quantity of the commodity sold, the quantity of commodity transaction failure of described network shop, described network shop and the quantity of the unfavorable ratings information received by the quantity of dispute, described network shop appears in buyer, described network shop receives the quantity of anonymous ratings, the quantity of the anonymous deal of described network shop, the wash sale of described network shop quantity at least one information.

Optionally, described second recommendation information comprises M bar and recommends sub-information, and M is positive integer, and every bar recommends the negative Transaction Information of the corresponding class of sub-information, in described acquisition and when exporting the second recommendation information, described method also comprises:

Judge that described M bar recommends in sub-information i-th to recommend the quantity comprised in sub-information whether to be greater than the i-th predetermined threshold value, i is the integer of 1 to M, i-th quantity of recommending the quantity comprised in sub-information to characterize the negative Transaction Information of the i-th class;

When recommending the quantity comprised in sub-information to be greater than described i-th predetermined threshold value for described i-th, produce corresponding information.

Second aspect, the embodiment of the present invention provides a kind of signal conditioning package, comprising:

First judge module, for when the accessing operation of access first webpage being detected, judges that whether described first webpage is the webpage of preset kind;

Obtain module, for when described first webpage is the webpage of described preset kind, obtain the descriptor of the corresponding destination object of described first webpage;

, be there is for being determined by described descriptor the first recommendation information associated with described destination object, and export described first recommendation information in determination module.

Optionally, the webpage of described preset kind is specially: shopping class webpage; Described acquisition module, specifically for:

Described determination module, specifically for:

Optionally, in described descriptor be: during the buyer's guide information of at least one commodity described, described determination module, specifically comprises:

First determining unit, for determining the shop merchandise classification information of the commodity that described network shop is sold by the buyer's guide information of at least one commodity described;

First search unit, determines described first recommendation information, the recommendation information that described first recommendation information is corresponding with first network link information for carrying out search using described shop merchandise classification information as keyword.

Optionally, described first search unit, specifically for:

Optionally, be specially in described descriptor: during the evaluation information of described first commodity, described determination module, specifically comprises:

Second determining unit, determines the first keyword set for carrying out analysis to the evaluation information of described first commodity;

3rd determining unit, for determining the keyword of the first kind from described first keyword set, the keyword of the described first kind is specially: carry out the keyword of unfavorable ratings to described first commodity or described first commodity carried out to the keyword of front evaluation, the keyword of the described first kind is described first recommendation information.

Optionally, described 3rd determining unit, specifically comprises:

First determines subelement, for determining the type of merchandise information of described first commodity;

Second determines subelement, for the type of merchandise information based on described first commodity, from the keywords database of multiple type of merchandise, determines the first keywords database;

First judgment sub-unit, for judging in described first keyword set, whether each keyword is positioned at described first keywords database one by one;

3rd determines subelement, when being positioned at described first keywords database for a certain keyword in the first keyword set, then determines that it is the keyword of the described first kind.

Optionally, be specially in described descriptor: during the buyer's guide information of described first commodity, described determination module, specifically comprises:

4th determining unit, for determining the type of merchandise information of described first commodity and the brand message of described first commodity by the buyer's guide information of described first commodity;

Second search unit, for using the brand message of the type of merchandise information of described first commodity and described first commodity as keyword, search obtains described first recommendation information, and described first recommendation information is: the recommendation information corresponding with second network link information.

Optionally, described second search unit, specifically for:

Optionally, described determination module, specifically comprises:

Acquiring unit, for obtaining out the second keyword set from described descriptor;

Judging unit, for judging whether a certain keyword in described second keyword set is arranged in a certain predetermined keyword list of multiple predetermined keyword list one by one;

5th determining unit, when being positioned at the first predetermined keyword list for the first keyword in described second keyword set, determine described first recommendation information comprising fisrt feature information, described fisrt feature information is corresponding with described first predetermined keyword list.

Optionally, described 5th determining unit, specifically comprises:

4th determines subelement, for when described first keyword is positioned at described first predetermined keyword list, determines the number percent of keyword shared by described second keyword set in described first predetermined keyword list;

Second judgment sub-unit, for judging whether described number percent is greater than predetermined threshold value;

Add subelement, for when described number percent is greater than described predetermined threshold value, described fisrt feature information is added described first recommendation information.

Optionally, described device also comprises:

Acquisition module, for when described first webpage is described shopping class webpage, obtains and export the second recommendation information, described second recommendation information is specially: the negative Transaction Information of described network shop.

Optionally, described second recommendation information comprises M bar and recommends sub-information, and M is positive integer, and every bar recommends the negative Transaction Information of the corresponding class of sub-information, and described device also comprises:

Second judge module, for when obtaining and export the second recommendation information, judge that described M bar recommends in sub-information i-th to recommend the quantity comprised in sub-information whether to be greater than the i-th predetermined threshold value, i is the integer of 1 to M, i-th quantity of recommending the quantity comprised in sub-information to characterize the negative Transaction Information of the i-th class;

Reminding module, for when recommending the quantity comprised in sub-information to be greater than described i-th predetermined threshold value for described i-th, produces corresponding information.

Beneficial effect of the present invention is as follows:

In the embodiment of the present application, disclose a kind of information processing method, comprising: when the accessing operation of access first webpage being detected, judge that whether the first webpage is the webpage of preset kind; When the first webpage is the webpage of preset kind, obtain the descriptor of the corresponding destination object of the first webpage; Determined to there is the first recommendation information associated with destination object by descriptor, and export the first recommendation information.Due to the descriptor of the destination object of the first webpage can be obtained and then analyze the demand of user, then the first recommendation information associated with the existence of destination object is obtained, and do not need user to access a large amount of pages, user is not needed to carry out induction-arrangement yet, so reach the technique effect of reduction equipment from the data volume of Network Capture data, and then reduce data transmission burden, and reduce by the network platform do shopping consuming time, improve user experience.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of information processing method in the embodiment of the present invention;

Fig. 2 is the process flow diagram determining to exist with the commodity of network shop or network shop the first kind of way of the first recommendation information associated in embodiment of the present invention information processing method by descriptor;

Fig. 3 is the process flow diagram determining to exist with the commodity of network shop or network shop the second way of the first recommendation information associated in embodiment of the present invention information processing method by descriptor;

Fig. 4 is the process flow diagram of the keyword determining the first kind in embodiment of the present invention information processing method from the first keyword set;

Fig. 5 is the process flow diagram determining to exist with the commodity of network shop or network shop the third mode of the first recommendation information associated in embodiment of the present invention information processing method by descriptor;

Fig. 6 is the process flow diagram determining to exist with the commodity of network shop or network shop the 4th kind of mode of the first recommendation information associated in embodiment of the present invention information processing method by descriptor;

Fig. 7 is the process flow diagram of the first recommendation information determining to comprise fisrt feature information in embodiment of the present invention information processing method;

Fig. 8 is the process flow diagram producing information in embodiment of the present invention information processing method when obtaining and export the second recommendation information;

Fig. 9 is the schematic diagram of the second recommendation information in embodiment of the present invention information processing method;

Figure 10 is the structural drawing of signal conditioning package in the embodiment of the present invention.

Embodiment

The technical scheme of the embodiment of the present application is for solving the problems of the technologies described above, and general thought is as follows:

Provide a kind of information processing method, comprising: when the accessing operation of access first webpage being detected, judge that whether the first webpage is the webpage of preset kind; When the first webpage is the webpage of preset kind, obtain the descriptor of the corresponding destination object of the first webpage; Determined to there is the first recommendation information associated with destination object by descriptor, and export the first recommendation information.Due to the descriptor of the destination object of the first webpage can be obtained and then analyze the demand of user, then the first recommendation information associated with the existence of destination object is obtained, and do not need user to access a large amount of pages, user is not needed to carry out induction-arrangement yet, so reach the technique effect of reduction equipment from the data volume of Network Capture data, and then reduce data transmission burden, and reduce by the network platform do shopping consuming time, improve user experience.

In order to better understand technique scheme, below by accompanying drawing and specific embodiment, technical solution of the present invention is described in detail, the specific features being to be understood that in the embodiment of the present invention and embodiment is the detailed description to technical solution of the present invention, instead of the restriction to technical solution of the present invention, when not conflicting, the technical characteristic in the embodiment of the present invention and embodiment can combine mutually.

First aspect, the embodiment of the present invention provides a kind of information processing method, please refer to Fig. 1, comprising:

Step S101: when the accessing operation of access first webpage being detected, judges that whether the first webpage is the webpage of preset kind;

Step S102: when the first webpage is the webpage of preset kind, obtains the descriptor of the corresponding destination object of the first webpage;

Step S103: determined to there is the first recommendation information associated with destination object by descriptor, and export the first recommendation information.

In step S101, the webpage of preset kind is such as: shopping class webpage, books read class webpage, teaching class webpage etc., in specific implementation process, can be prestored the multiple domain names corresponding to webpage of preset kind, then the domain name of the first webpage is mated with the domain name prestored, if in the domain name prestored that the domain name of the first webpage comprises, then determine that the first webpage is the webpage of preset kind; For the webpage of preset kind for shopping class webpage, can be prestored the domain name of multiple shopping webpage, after accessing operation access first webpage being detected, obtain the reference address of the first webpage, then judge whether the domain name of the reference address of the first webpage is included in the domain name of the multiple shopping websites prestored, if the domain name of the reference address of the first webpage is included in the domain name of the multiple shopping websites prestored, then determine that the first webpage is for shopping class webpage, otherwise determine that the first webpage is not shopping class webpage.

In step S102, based on the difference of the webpage of preset kind, the descriptor of the destination object corresponding to the first webpage of acquisition is also different, and to enumerate wherein below several is introduced, and certainly, in specific implementation process, is not limited to following several situation.

The first, if the webpage of preset kind is shopping class webpage, obtain the descriptor of the corresponding destination object of the first webpage, be specially: the descriptor obtaining at least one commodity of the network shop corresponding to the first webpage, at least one commodity is destination object;

For example, when the first webpage is for shopping class webpage, the first webpage is such as: sales page of the homepage of network shop, the publicity page of network shop, the first commodity etc.

In this case, the descriptor of at least one commodity can be divided into polytype descriptor, and two kinds of enumerating below are wherein introduced, and certainly, in specific implementation process, is not limited to following two kinds of situations.

1) descriptor is specially: the buyer's guide information of at least one commodity.

For example, if the first webpage is the homepage of network shop, then the first webpage can show the hot item of this network shop multiple, and at least one commodity can hot item shown by the first webpage; Or at least one commodity is all commodity of this network shop.For each commodity, network shop all can add buyer's guide information for it, to enable buyer understand corresponding commodity, carry out analysis and arrangement by the buyer's guide information obtaining at least one commodity, just can roughly understand the commodity that this network shop or network shop are sold.

As further preferred embodiment, when first page is the sales page of the first commodity, descriptor is specially: the buyer's guide information of the first commodity.

For example, if current page is the sales page of shoes, the information such as price, size, brand, style of these shoes then can be shown in the sales page of these shoes, these information are just descriptor, because user is when browsing the sales page of the first commodity, wish purchase first commodity often, so in this case, obtain the first recommendation information by the descriptor of the first commodity, then the first recommendation information obtained more targetedly and more accurate.

2) descriptor is specially: user is to the evaluation information of at least one commodity.

For example, multiple commodity can be there are in the first webpage place network shop, user is after these multiple commodity of purchase, evaluation information can be produced for each commodity, the information such as the quality of commodity, the attitude of hotel owner can be found out, so the evaluation information that can obtain at least one commodity is as descriptor from these evaluation informations.

As further preferred embodiment, when first page is the sales page of the first commodity, descriptor is specially: user is to the evaluation information of the first commodity.What browse due to user is the sales page of the first commodity, so illustrate that user has a mind purchase first commodity, in this case, the evaluation information based on the first commodity can obtain comparatively accurate first recommendation information.

The second, if the webpage of preset kind is books read class webpage, then obtain the descriptor of the corresponding destination object of the first webpage, be specially: the descriptor obtaining the books of the first webpage, the books of the first webpage are destination object, and wherein the descriptor of the books of the first webpage is such as: summary info, review information etc.

The third, if the webpage of preset kind is teaching class webpage, then obtain the descriptor of the destination object corresponding to the first webpage, be specially: the descriptor obtaining the instructional video of the first webpage, the instructional video of the first webpage is destination object, and the descriptor of the instructional video of the first webpage is such as: the content introduction information, directory information, review information etc. of the instructional video of the first webpage.

In step S103, based on the difference of the webpage of preset kind, the mode obtaining the first recommendation information is also different, is introduced to enumerate wherein several below, certainly, in specific implementation process, is not limited to following several situation.

The first, when the webpage of preset kind is for shopping class webpage, determines to there is the first recommendation information associated with destination object by descriptor in step S103, be specially:

Determine to there is with the commodity of network shop or network shop the first recommendation information associated by descriptor.

In this case, the mode obtaining the first recommendation information can also be subdivided into multiple different mode, is introduced to enumerate wherein several below, certainly, in specific implementation process, is not limited to following several situation.

1) in descriptor be: during the buyer's guide information of at least one commodity, determine to there is with the commodity of network shop or network shop the first recommendation information associated by descriptor, please refer to Fig. 2, specifically comprise:

Step S201: the shop merchandise classification information of the commodity sold by the buyer's guide information determination network shop of at least one commodity;

Step S202: carry out search using shop merchandise classification information as keyword and determine the first recommendation information, the recommendation information that the first recommendation information is corresponding with first network link information.

In step S201, some network shops may only sell a kind of commodity, some network shops may sell multiple commodity, the title of commodity can be extracted from the buyer's guide information of commodity, then the merchandise classification information of each commodity is judged one by one, and then the commodity of network shop are divided into multiclass, such as shown in table 1:

Table 1

Merchandise classification information	Footwear	Jacket	Skirt
				Kind	20	4	4

Then, in merchandise classification information, the rank of kind is positioned at front N (such as: merchandise classification information 1) as shop merchandise classification information, such as, footwear in table 1.

In step S202, the first recommendation information can be the recommendation information of various ways, and three kinds of enumerating below are wherein introduced, and certainly, in specific implementation process, is not limited to following three kinds of situations.

1. the first recommendation information is specially: the keyword of first network link information and first network link information.

For example, can search for using " footwear " as keyword, and then determine the first network link information relevant to " footwear ", then obtain and export keyword and the first network link information of first network link information, wherein, the keyword of first network link information is such as: the title, summary info etc. of the first network link information place page, such as, can return keyword and the first network link information of first network link information as shown in table 2:

Table 2

Keyword	First network link information
		How to choose footwear	http://xxxx.com
The most how history identifies the false sport footwear attack strategy of Taobao	http://yyyy.com

Then, if how the user of net purchase to identifying that the false sport footwear of Taobao is interested, then can clickthrough http://yyyy.com, and then jump to the corresponding page and check relevant information.

Due in such scheme, first recommendation information is first network link information, the data of first network link information are not namely needed all to obtain, but based on the page of the subnetwork link information in the selection operational access first network link information of user, so reach the technique effect reducing data transmission burden.

2. the first recommendation information is specially: the content of pages corresponding to first network link information.

For example, namely can " footwear " search for as keyword, and then determine the first network link information relevant to " footwear ", then the content of pages of first network link information is obtained as the first recommendation information, wherein, if the content of pages of first network link information is less, then obtain whole content of pages of first network link information as the first recommendation information, if the content of pages of first network link information is more, then obtain the partial page content of first network link information as the first recommendation information.

3. the first recommendation information is specially: the information obtained after summarizing the content of pages of first network link information.

For example, the unwanted content of a lot of user may be comprised in the content of pages of first network link information, such as: the model for " how choosing footwear " may comprise the experience that a lot of building-owner buys footwear, a lot of building-owner to have bought false footwear harrowing experience in Taobao may be comprised for " the most how history identifying the false sport footwear attack strategy of Taobao ", even and part relevant to title in the content of pages of first network link information, also may be very loaded down with trivial details, so can semantic analysis be passed through, first network link information is summarized, and then determine that the information relevant and brief and concise to keyword is as the first recommendation information.

By such scheme, reduce data transmission burden on the one hand, decrease the screening time of net purchase user to useful information on the other hand, thus improve net purchase efficiency.

In step S202, when acquisition the first recommendation information, can multitude of different ways be divided into again, introduce two kinds of modes wherein below.

1. carry out search using shop merchandise classification information as keyword and determine the first recommendation information, be specially:

From the first corresponding relation prestored, obtain the first recommendation information by shop merchandise classification information searching, in the first corresponding relation, include the corresponding relation of keyword and recommendation information;

For example, first can determine multiple shops merchandise classification information, for each shop merchandise classification information in this multiple shops merchandise classification information, the recommendation information of its correspondence can be determined in the following manner: search for using shop merchandise classification information as keyword, obtain multiple network link information, then filter out from this multiple network link information the higher network link information of quality (such as: access times are more, point praise more, comment on more etc.); So can set up corresponding relation between shop merchandise classification information and network link information as the first corresponding relation; Also the content of pages of shop merchandise classification information and network link information can be set up as the first corresponding relation; Also can first summarize the page of network link information, the information obtained after then setting up shop merchandise classification information and summarizing the content of pages of network link information is as the first corresponding relation.

Further, before filter out the higher network link information of quality from multiple network link information, the enquirement that can also first carry out from Network Capture for shop merchandise classification information, then search is putd question to obtain multiple network link information by these, due under normal circumstances, user is often when shopping, shop merchandise classification information for correspondence is putd question to, so user can be made when buying the commodity corresponding to the merchandise classification information of shop by which, do not need to be putd question to by network, just can obtain corresponding answer, so reach determine the first corresponding relation technique effect more accurately, and then determined first recommendation information is also more accurate, and reduce Internet Transmission burden.

In addition, in specific implementation process, when obtaining multiple webpage link information, the network link information of the whole network can be obtained, also can manually obtain reliable network link information by user, and when screening multiple webpage link information, can by manually screening, also can be screened by large data analysis, when being screened by large data analysis, several indexs of network link information can be chosen (such as: reply number, access number, point praises number etc.), then for each setup measures weight, then by weight, computing is summed up to these indexs, just can determine the score value of network link information, finally obtain the network link information that score value is positioned at former and set up corresponding relation with corresponding shop merchandise classification information.

By such scheme, the speed of acquisition first recommendation information can be improved.

2. carry out search using shop merchandise classification information as keyword and determine the first recommendation information, be specially:

Carry out search using shop merchandise classification information as keyword in a network and obtain first network link information, and determine the first recommendation information by first network link information.

For example, namely after determining shop merchandise classification information, directly searched for by shop merchandise classification information, and then obtain multiple network link information, then by large data analysis, this multiple network link information is screened, when being screened by large data analysis, several indexs of network link information can be chosen (such as: reply number, access number, point praises number etc.), then for each setup measures weight, then by weight, computing is summed up to these indexs, just can determine the score value of network link information, finally obtain score value and be positioned at the network link information of former as first network link information.

If the first recommendation information is first network link information, then directly export first network link information; If the first recommendation information is the content of pages of first network link information, then access first network link information, then obtain corresponding content of pages; If the information that the content of pages that the first recommendation information is first network link information obtains after summarizing, then first access the content of pages of first network link information, then induction-arrangement carried out to the content of pages of first network link information and then obtain the first recommendation information.

By such scheme, reach the technique effect reducing and store burden.

2) be specially in descriptor: during the evaluation information of the first commodity, determined to there is the first recommendation information associated with the commodity of network shop by descriptor, please refer to Fig. 3, specifically comprise:

Step S301: analysis is carried out to the evaluation information of the first commodity and determines the first keyword set;

Step S302: the keyword determining the first kind from the first keyword set, the keyword of the first kind is specially: carry out the keyword of unfavorable ratings to the first commodity or the first commodity carried out to the keyword of front evaluation, the keyword of the first kind is the first recommendation information.

In step S301, keyword extraction can be carried out to the evaluation information of the first commodity, and then obtain multiple keyword, and the number of times that each keyword occurs, the number of times that this multiple keyword and each keyword occur is composition the first keyword set just; Or, semantic analysis can be carried out to the evaluation information of the first commodity, and then determine the frequency that multiple keyword and each keyword occur.

In step S302, the dictionary of the keyword of a first kind can be set, then the keyword in the dictionary of the keyword of each keyword in the first keyword set and the first kind is mated, to determine whether this keyword is positioned at the dictionary of the keyword of the first kind, if this keyword is positioned at the dictionary of the keyword of the first kind, then determine that this keyword is the keyword of the first kind.

Wherein, the difference can searching for whole shopping at network platform is commented, then difference is commented and carry out keyword extraction, and obtain quantity be positioned at before the keyword of (such as: first 1000, first 2000 etc.) add the dictionary of the keyword of the first kind, and then the keyword of the determined first kind is the keyword the first commodity being carried out to unfavorable ratings; Or, the favorable comment of whole shopping at network platform can be searched for, then keyword extraction is carried out to favorable comment, and obtain quantity be positioned at before the keyword of (such as: first 1000, front 2000 etc.) add the dictionary of the keyword of the first kind, and then the keyword of the determined first kind is the keyword the first commodity being carried out to front evaluation.

In addition, in specific implementation process, both the dictionary of keyword commodity being carried out to front evaluation can also be set up, set up again the dictionary of keyword commodity being carried out to unfavorable ratings, then the keyword in these two dictionaries is mated, if certain keyword had both been positioned at the dictionary of keyword commodity being carried out to front evaluation, be positioned at again the dictionary of keyword commodity being carried out to unfavorable ratings, then this keyword is removed from the keyword of the first kind, do not show as the first recommendation information.By such scheme, reach can obtain accurately user to the first commodity carry out negative or or the data evaluated of front, and do not need user to screen when net purchase, so further improve the efficiency of net purchase.

As further preferred embodiment, in step S302, from the first keyword set, determine the keyword of the first kind, please refer to Fig. 4, specifically comprise:

Step S401: the type of merchandise information determining the first commodity;

Step S402: based on the type of merchandise information of the first commodity, determines the first keywords database from the keywords database of multiple type of merchandise information;

Step S403: judge in the first keyword set, whether each keyword is positioned at the first keywords database one by one;

Step S404: when a certain keyword in the first keyword set is positioned at the first keywords database, then determine that it is the keyword of the first kind.

In step S401, type of merchandise information is such as: footwear, jacket, skirt etc., can be determined the type of merchandise information of the first commodity by the descriptor of the first commodity.

In step S402, the evaluation information of whole shopping at network platform can be searched in advance for each type of merchandise, then determine the keywords database (wherein keywords database can also be divided into front to evaluate word bank and unfavorable ratings word bank) of corresponding goods type information, and then set up the corresponding relation of multiple type of merchandise and keywords database; Thus the type of merchandise information of direct first commodity can determine the first keywords database.

In step S404, if the first keywords database is the keyword type of merchandise information of the first commodity being carried out to unfavorable ratings, a certain keyword is positioned at the first keywords database, then directly determine that this keyword is the keyword (also namely the first kind is unfavorable ratings type) of unfavorable ratings; If the first keywords database is the keyword type of merchandise information of the first commodity being carried out to front evaluation, a certain keyword is positioned at the first keywords database, then directly determine that this keyword is the keyword (also namely the first kind is that type is evaluated in front) that front is evaluated;

If the first keywords database comprises evaluate word bank to the front of unfavorable ratings word bank and front evaluation that the type of merchandise information of the first commodity carries out unfavorable ratings, if then a certain keyword is arranged in the unfavorable ratings word bank of the first keywords database, then determine that the first kind is: unfavorable ratings type; If word bank is evaluated in the front that certain keyword is arranged in the first keywords database, then determine that the first kind is: type is evaluated in front.

Due to different keywords databases can be there is for different type of merchandise information, and then determine the first corresponding recommendation information, so have obtained recommendation information technique effect more accurately based on this keywords database.

In addition, using the keyword of the first kind as after the first recommendation information is supplied to user, can also screen the first keyword based on the operation of the selection of user, finally determined first keywords database is through the garbled keyword dictionary of user, can be optimized the first keywords database by the program, and then finally determine that the first recommendation information is more accurate.

Such as, the keyword of the first kind is the keyword of unfavorable ratings, by the keyword in the first keyword set is mated with the first keywords database, determine following keyword: sound is large, damage, weak effect, and the first commodity are sound equipment, a lot of user feeds back and thinks: sound does not include unfavorable ratings greatly, so this keyword large for sound can be removed from the first dictionary.

3) be specially in descriptor: during the buyer's guide information of the first commodity, determined to there is the first recommendation information associated with the commodity of network shop by descriptor, please refer to Fig. 5, specifically comprise:

Step S501: determine the type of merchandise information of the first commodity and the brand message of the first commodity by the buyer's guide information of the first commodity;

Step S502: using the brand message of the type of merchandise information of the first commodity and the first commodity as keyword, search acquisition first recommendation information, the first recommendation information is: the recommendation information corresponding with second network link information.

In step S501, just can determine the type of merchandise information (such as: footwear) of the first commodity by carrying out keyword extraction to the buyer's guide information of the first commodity, and the brand message of the first commodity (such as: Nike).

In step S502, the first recommendation information is specially: second network link information; Or the content of pages corresponding to second network link information; Or the information obtained after the content of pages of second network link information summarized.

Can obtain the first recommendation information in several ways in step S502, two kinds of enumerating below are wherein introduced, and certainly, in specific implementation process, are not limited to following two kinds of situations.

1. obtain the first recommendation information using the brand message of the type of merchandise information of the first commodity and the first commodity as keyword, specifically comprise:

From the second corresponding relation prestored, obtain the first recommendation information by the type of merchandise information of the first commodity and the brand message of the first commodity as keyword lookup, in the second corresponding relation, include the corresponding relation of keyword and recommendation information.

Wherein, the second corresponding relation and the first corresponding relation similar, the mode obtaining the first recommendation information by such scheme is similar with the mode being obtained the first recommendation information by the first corresponding relation, so do not repeat them here.

2. obtain the first recommendation information using the brand message of the type of merchandise information of the first commodity and the first commodity as keyword, specifically comprise:

Carry out search acquisition second network link information using the brand message of the type of merchandise information of the first commodity and the first commodity in a network as keyword, and determine the first recommendation information by second network link information.

The mode of acquisition second network link information is searched in a network with similar by the mode of shop merchandise classification information acquisition first network link information, so do not repeat them here by such scheme.

4) determine to there is with the commodity of network shop or network shop the first recommendation information associated by descriptor, please refer to Fig. 6, specifically comprise:

Step S601: obtain out the second keyword set from descriptor;

Step S602: judge in the second keyword set, whether a certain keyword is arranged in a certain lists of keywords of multiple predetermined keyword list one by one;

Step S603: when the first keyword in the second keyword set is positioned at the first predetermined keyword list, determine the first recommendation information comprising fisrt feature information, fisrt feature information is corresponding with predetermined keyword list.

In step S601, if the first webpage is the homepage of network shop, then now do not point to a certain particular commodity, so can determine that at least one information in the buyer's guide information of all commodity of the homepage of network shop or evaluation information is as descriptor, if the first webpage is the sales page of the first commodity, then now the first webpage points to the first commodity, so at least one information can got in the buyer's guide information of the first commodity or evaluation information is as descriptor, but may evaluation information do not had for the first commodity, buyer's guide information only based on the first commodity can not comprehensively understand the first commodity, and all there is association with the first commodity in the commodity of network shop, so when the first webpage is the sales page of the first commodity, also the buyer's guide information of the commodity of whole network shop and evaluation information can be obtained as descriptor.

Can pass through keyword extraction or semantic analysis in step S601, and then determine multiple keyword, this multiple keyword then forms the second keyword set.

In step S602, multiple characteristic informations of the commodity of network shop or network shop can be pre-determined, for its characteristic information of network shop be such as: false propaganda, pass off etc.; For the commodity of network shop, its characteristic information is such as: size is not to, poor quality etc.

Then by large data analysis, determining the predetermined keyword corresponding to each characteristic information, take characteristic information as false propaganda is example, and its predetermined keyword is such as: false one pay for ten, extremely cheap price etc.; Take characteristic information as pass off be example, its predetermined keyword is such as: fake products, be not certified products etc.; With characteristic information be size not to being example, its predetermined keyword is such as: bigger than normal, less than normal, misfit etc.; Take characteristic information as poor quality be example, its predetermined keyword is such as: off-line, have the end of a thread, coarse etc.

And then after determining the second keyword set, just each keyword in the second keyword set is mated with each predetermined keyword list in predetermined keyword list one by one.

In step S603, if determine that the keyword (such as: be not certified products) in the second keyword set is positioned at a certain predetermined keyword list (such as: the predetermined keyword list corresponding to the characteristic information of pass off), then characteristic of correspondence information (such as: pass off) is added the first recommendation information; If further determine that the keyword (such as: false pays for ten) in the second keyword set is positioned at another predetermined keyword list (such as: the predetermined keyword list corresponding to the characteristic information of false propaganda), then this characteristic information (such as: false propaganda) is added the first recommendation information, and then finally determined first recommendation information is: pass off false propaganda.

As further preferred embodiment, when the first keyword in the second keyword set is positioned at the first predetermined keyword list, determines the first recommendation information comprising fisrt feature information, please refer to Fig. 7, specifically comprise:

Step S701: when the first keyword is positioned at the first predetermined keyword list, determines the number percent of keyword shared by the second keyword set in the first predetermined keyword list;

Step S702: judge whether number percent is greater than predetermined threshold value;

Step S703: when number percent is greater than predetermined threshold value, adds the first recommendation information by fisrt feature information.

In step S701, when the first keyword is positioned at the first predetermined keyword list, then illustrate that the commodity for network shop or network shop exist corresponding description, but do not represent that this description must be accurately, such as: in certain network shop, include several thousand evaluations, wherein have several to evaluate corresponding keywords to be: fake products, in this case, possible buyer assert that corresponding goods be the comment of fake products is wrong, so in order to prevent in this case, then need the number percent that the keyword in the list of judgement first predetermined keyword in the first predetermined keyword list is shared in the second keyword set.

For example, after determining that the first keyword (such as: fake products) is positioned at the first predetermined keyword list, just determine the keyword (such as: fake products is not certified products) comprised in the first predetermined keyword list, then determine the quantity of these two keywords comprised in the second keyword set, the quantity finally by these two keywords just can calculate number percent divided by the sum of the keyword comprised in the second keyword set.

In step S702, can different predetermined threshold value be set according to the actual requirements, for different predetermined keyword lists, also different predetermined threshold value can be set, such as: for the corresponding keyword of recommended information in " false one pays for ten " this shop, it is lower that its predetermined threshold value can be arranged, because substantially there is this word, then illustrate that the probability of shop pass off is higher; For the keyword that " fake products " this user evaluates, it is higher a little that its predetermined threshold value can be arranged, because just have reference value when the evaluation of user reaches a certain amount of.

In step S703, if number percent is greater than predetermined threshold value, then illustrate that the probability that the keyword in the first predetermined keyword list corresponding to fisrt feature information occurs in the second keyword set is higher, so in this case, determine that fisrt feature information is the characteristic information that of the commodity of network shop or network shop is comparatively correct, so added the first recommendation information; Otherwise, think that fisrt feature information is not too accurate, so do not added the first recommendation information.

By such scheme, reach determined fisrt feature information technique effect more accurately.

The second, if the webpage of preset kind is books read class webpage, determines to there is the first recommendation information associated with destination object by descriptor in step S103, be specially:

Determined to there is the first recommendation information associated with the books of the first webpage by descriptor.

For example, suppose that descriptor is the summary info of the books of the first webpage, then can by the overall evaluation information of the books of Network Capture first webpage, the first recommendation information is it can be used as to be supplied to user, like this, user does not need the whole network search just can obtain the evaluation information of the books to the first webpage, and then determines whether these books are worth seeing, thus has the technique effect reduced from the data volume of Network Capture.

The third, if the webpage of preset kind is teaching class webpage, determines to there is the first recommendation information associated with destination object by descriptor in step S103, be specially:

There is the first recommendation information associated in the instructional video being determined at the first webpage by descriptor.

For example, suppose that descriptor is the content introduction information of the instructional video of the first webpage, then by this content introduction information, can obtain identical with the content of this instructional video and evaluate other instructional video better as the first recommendation information, and then be supplied to user, in this case, can ensure that user obtains more excellent resource;

Suppose that descriptor is the review information of the instructional video of the first webpage, the keyword (such as: sound is less, detailed not etc.) of keyword (such as: clear logic, clear and easy to understand etc.) and unfavorable ratings is evaluated in the front that then can extract from this review information for this instructional video, then keyword is evaluated in front and unfavorable ratings keyword is supplied to user as the first recommendation information, and then enable user understand the Pros and Cons of the instructional video of the first webpage easily, to determine whether to need to learn this instructional video.

As further preferred embodiment, method also comprises:

When the first webpage is for shopping class webpage, obtain and export the second recommendation information, the second recommendation information is specially: the negative Transaction Information of network shop.

In addition, under normal circumstances, the Recent Activity of network shop has reference value more, so the negative Transaction Information of network shop can only can comprise the negative Transaction Information of the network shop within a preset time period, preset time period is such as: 2 months, 30 days, 15 days etc.

By such scheme, reach and can provide the technique effect of reference informations much more more for shopping at network, thus further save the time of shopping at network.And, the usual entrance of second recommendation information is comparatively hidden, if user oneself searches, the multiple pages opening network shop may be needed to search, in this case, also Internet Transmission is caused to be born heavier, so can also reach by above-mentioned the technique effect further reducing Internet Transmission burden.

In specific implementation process, the negative Transaction Information of network shop can comprise much information, such as: comprise at least one information in several information below, certainly, in specific implementation process, the negative Transaction Information of network shop is not limited to following several situation.

The first, negative Transaction Information specifically comprises: the commodity that network shop is sold are by the quantity of reimbursement.

For example, the commodity that network shop is sold can be by the number of times of reimbursement by the quantity of reimbursement, such as: reimbursement 17 times; Commodity also can be a number percent by the quantity of reimbursement, such as: reimbursement rate is 5.30%, can certainly both comprise, such as: reimbursement rate is 5.30%, reimbursement number of times 17 times, or nearly 30 days reimbursement rates are 5.30%, reimbursement number of times 17 times.

The second, negative Transaction Information specifically comprises: the quantity that the commodity sold of network shop are complained;

For example, the complained quantity of the commodity sold of network shop can for complained number of times, such as: reimbursement 2 times; The complained quantity of commodity also can be a number percent, such as: reimbursement rate is 0.50%, can certainly both comprise, such as: the rate of complaints is 0.50%, complain number of times 2 times, or nearly 30 days the rate of complaints is 0.50%, complains number of times 2 times.

The third, negative Transaction Information specifically comprises: the quantity of the commodity transaction failure of network shop;

For example, the quantity of Fail Transaction can be the number of times of Fail Transaction, such as: Fail Transaction 500 times; The quantity of Fail Transaction also can be a number percent, such as: Fail Transaction ratio: 90%, can certainly both comprise, such as: Fail Transaction ratio is 0.50%, Fail Transaction number of times is 500 times, or nearly 30 days Fail Transaction ratios are 0.50%, and Fail Transaction number of times is 500 times.

4th kind, negative Transaction Information specifically comprises: the quantity of dispute appears in network shop and buyer;

For example, network shop and buyer occur that the quantity of dispute can for occurring the number of times of dispute, such as: occur dispute 0 time; The quantity of appearance dispute also can be a number percent, such as: occur dispute ratio: 0%, can certainly both comprise, such as: occur that the number percent of dispute is 0%, the number of times of appearance dispute is 0 time, or nearly number percent occurring dispute for 30 days is 0%, occurs that the number of times of dispute is 0 time.

5th kind, negative Transaction Information specifically comprises: the quantity of the unfavorable ratings information received by network shop;

For example, unfavorable ratings information is such as: the difference of network shop is commented, the quantity of unfavorable ratings information can be a numerical value, such as: difference comments number to be 7, also can be number percent, such as: difference comments shared ratio to be: 3%, can certainly both comprise, such as: the number percent that difference is commented is 3%, the quantity that difference is commented is 7, or the number percent that nearly 30 days differences are commented is 3%, the quantity that difference is commented is 7.

6th kind, negative Transaction Information specifically comprises: the quantity of the anonymous ratings that network shop receives;

For example, the quantity of anonymous ratings can be the number of times of anonymous ratings, such as: anonymous ratings number of times is 600 times, the quantity of anonymous ratings also can be a number percent, such as: the number percent of anonymous ratings is: 100%, can certainly both comprise, such as: the number percent of anonymous ratings is: 100%, anonymous ratings number of times is 600 times, or the number percent of nearly 30 days anonymous ratings is: 100%, and anonymous ratings number of times is 600 times.

7th kind, negative Transaction Information specifically comprises: the quantity of the anonymous deal of network shop.

For example, the quantity of anonymous deal can be the number of times of anonymous deal, such as: anonymous deal number of times is 1000 times, the quantity of anonymous deal also can be a number percent, such as: the number percent of anonymous deal is: 100%, can certainly both comprise, such as: the number percent of anonymous deal is: 100%, anonymous deal number of times is 1000 times, or the number percent of nearly 30 days anonymous deals is: 100%, and anonymous deal number of times is 1000 times.

8th kind, negative Transaction Information specifically comprises: the quantity of the wash sale of network shop.

For example, the quantity of wash sale can be the number of times of wash sale, such as: wash sale number of times is 2 times, the quantity of wash sale also can be a number percent, such as: the number percent of wash sale is: 1%, can certainly both comprise, such as: the number percent of wash sale is: 1%, wash sale number of times is 2 times, or the number percent of nearly 30 days wash sales is: 1%, and wash sale number of times is 2 times.

As further preferred embodiment, the second recommendation information comprises M bar and recommends sub-information, and M is positive integer, and every bar recommends the negative Transaction Information of the corresponding class of sub-information, and when obtaining and export the second recommendation information, please refer to Fig. 8, method also comprises:

Step S801: judge that M bar recommends in sub-information i-th to recommend the quantity comprised in sub-information whether to be greater than the i-th predetermined threshold value, i is the integer of 1 to M, i-th quantity of recommending the quantity comprised in sub-information to characterize the negative Transaction Information of the i-th class;

Step S802: when recommending the quantity comprised in sub-information to be greater than the i-th predetermined threshold value for i-th, produce corresponding information.

For example, suppose this M bar recommend sub-information and the i-th predetermined threshold value as shown in table 3:

Table 3

Sub-information recommended by i-th strip	Content	I-th predetermined threshold value
			Article 1	Reimbursement rate 5.30%	3％
Article 2	Difference comments percent 2%	3％
			Article 3	Anonymous buying rate 100%	30％
Article 4	Wash sale number of times 1 time	0％

As can be seen here, Article 1 recommends the quantity in sub-information to be greater than the first predetermined threshold value, so produce information based on this, such as: reimbursement rate is higher than industry average level; Article 2 recommends the quantity in sub-information to be less than the second predetermined threshold value, so do not produce information; Article 3 recommends the quantity in sub-information to be greater than the 3rd predetermined threshold value, so produce information, such as: buy without real name, please carefully check Bidder Information, guard against seller and brush list; Article 4 recommends the quantity in sub-information to be greater than the 4th predetermined threshold value, so produce information, such as: there is sell-fake-products situation etc.

In addition, can also show the opening time in current network shop in the second recommendation information, sell the parameters such as counterfeit goods number of times, as shown in Figure 9, be the schematic diagram of one second recommendation information.

Second aspect, based on same inventive concept, the embodiment of the present invention provides a kind of signal conditioning package, please refer to Figure 10, specifically comprises:

First judge module 100, for when the accessing operation of access first webpage being detected, judges that whether the first webpage is the webpage of preset kind;

Obtain module 110, for when the first webpage is the webpage of preset kind, obtain the descriptor of the corresponding destination object of the first webpage;

, be there is for being determined by descriptor the first recommendation information associated with destination object, and export the first recommendation information in determination module 120.

Optionally, the webpage of preset kind is specially: shopping class webpage; Obtain module 110, specifically for:

Obtain the descriptor of at least one commodity of the network shop corresponding to the first webpage, at least one commodity is destination object;

Determination module 120, specifically for:

Optionally, descriptor is specially: the buyer's guide information of at least one commodity; Or user is to the evaluation information of at least one commodity.

Optionally, when first page is the sales page of the first commodity, descriptor is specially: the buyer's guide information of the first commodity or user are to the evaluation information of the first commodity.

Optionally, in descriptor be: during the buyer's guide information of at least one commodity, determination module 120, specifically comprises:

First determining unit, for the shop merchandise classification information of commodity of being sold by the buyer's guide information determination network shop of at least one commodity;

First search unit, determines the first recommendation information, the recommendation information that the first recommendation information is corresponding with first network link information for carrying out search using shop merchandise classification information as keyword.

Optionally, the first recommendation information is specially: the keyword of first network link information and first network link information; Or the content of pages corresponding to first network link information; Or the information obtained after the content of pages of first network link information summarized.

Optionally, the first search unit, specifically for:

From the first corresponding relation prestored, obtain the first recommendation information by shop merchandise classification information searching, in the first corresponding relation, include the corresponding relation of keyword and recommendation information; Or

Optionally, be specially in descriptor: during the evaluation information of the first commodity, determination module 120, specifically comprises:

Second determining unit, determines the first keyword set for carrying out analysis to the evaluation information of the first commodity;

3rd determining unit, for determining the keyword of the first kind from the first keyword set, the keyword of the first kind is specially: carry out the keyword of unfavorable ratings to the first commodity or the first commodity carried out to the keyword of front evaluation, the keyword of the first kind is the first recommendation information.

Optionally, the 3rd determining unit, specifically comprises:

First determines subelement, for determining the type of merchandise information of the first commodity;

Second determines subelement, for the type of merchandise information based on the first commodity, from the keywords database of multiple type of merchandise, determines the first keywords database;

First judgment sub-unit, for judging in the first keyword set, whether each keyword is positioned at the first keywords database one by one;

3rd determines subelement, when being positioned at the first keywords database for a certain keyword in the first keyword set, then determines that it is the keyword of the first kind.

Optionally, be specially in descriptor: during the buyer's guide information of the first commodity, determination module, specifically comprises:

4th determining unit, for determining the type of merchandise information of the first commodity and the brand message of the first commodity by the buyer's guide information of the first commodity;

Second search unit, for using the brand message of the type of merchandise information of the first commodity and the first commodity as keyword, search acquisition first recommendation information, the first recommendation information is: the recommendation information corresponding with second network link information.

Optionally, the first recommendation information is specially: the keyword of second network link information and second network link information; Or the content of pages corresponding to second network link information; Or the information obtained after the content of pages of second network link information summarized.

Optionally, the second search unit, specifically for:

From the second corresponding relation prestored, obtain the first recommendation information by the type of merchandise information of the first commodity and the brand message of the first commodity as keyword lookup, in the second corresponding relation, include the corresponding relation of keyword and recommendation information; Or

Optionally, determination module 120, specifically comprises:

Acquiring unit, for obtaining out the second keyword set from descriptor;

Judging unit, for judging whether a certain keyword in the second keyword set is arranged in a certain predetermined keyword list of multiple predetermined keyword list one by one;

5th determining unit, when being positioned at the first predetermined keyword list for the first keyword in the second keyword set, determine the first recommendation information comprising fisrt feature information, fisrt feature information is corresponding with the first predetermined keyword list.

Optionally, the 5th determining unit, specifically comprises:

4th determines subelement, for when the first keyword is positioned at the first predetermined keyword list, determines the number percent of keyword shared by the second keyword set in the first predetermined keyword list;

Second judgment sub-unit, for judging whether number percent is greater than predetermined threshold value;

Add subelement, for when number percent is greater than predetermined threshold value, fisrt feature information is added the first recommendation information.

Optionally, device also comprises:

Acquisition module, for when the first webpage is for shopping class webpage, obtain and export the second recommendation information, the second recommendation information is specially: the negative Transaction Information of network shop.

Optionally, negative Transaction Information specifically comprises: the commodity sold of network shop by the quantity of reimbursement, network shop the quantity of anonymous ratings that the quantity of the unfavorable ratings information received by the quantity of dispute, network shop appears in the complained quantity of the commodity sold, the quantity of commodity transaction failure of network shop, network shop and buyer, network shop receives, the quantity of the anonymous deal of network shop, the wash sale of network shop quantity at least one information.

Optionally, the second recommendation information comprises M bar and recommends sub-information, and M is positive integer, and every bar recommends the negative Transaction Information of the corresponding class of sub-information, and device also comprises:

Second judge module, for when obtaining and export the second recommendation information, judge that M bar recommends in sub-information i-th to recommend the quantity comprised in sub-information whether to be greater than the i-th predetermined threshold value, i is the integer of 1 to M, i-th quantity of recommending the quantity comprised in sub-information to characterize the negative Transaction Information of the i-th class;

Reminding module, for when recommending the quantity comprised in sub-information to be greater than the i-th predetermined threshold value for i-th, produces corresponding information.

The one or more embodiment of the application, at least has following beneficial effect:

Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.And the present invention can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disk memory, CD-ROM, optical memory etc.) of computer usable program code.

The present invention describes with reference to according to the process flow diagram of the method for the embodiment of the present invention, equipment (system) and computer program and/or block scheme.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or square frame.These computer program instructions can being provided to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, making the instruction performed by the processor of computing machine or other programmable data processing device produce device for realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing device, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

These computer program instructions also can be loaded in computing machine or other programmable data processing device, make on computing machine or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computing machine or other programmable devices is provided for the step realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

Although describe the preferred embodiments of the present invention, those skilled in the art once obtain the basic creative concept of cicada, then can make other change and amendment to these embodiments.So claims are intended to be interpreted as comprising preferred embodiment and falling into all changes and the amendment of the scope of the invention.

Obviously, those skilled in the art can carry out various change and modification to the present invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.

Claims

1. an information processing method, is characterized in that, comprising:

2. the method for claim 1, is characterized in that, the webpage of described preset kind is specially: shopping class webpage; The descriptor of the corresponding destination object of described first webpage of described acquisition, is specially:

3. method as claimed in claim 2, it is characterized in that, described descriptor is specially: the buyer's guide information of at least one commodity described; Or user is to the evaluation information of at least one commodity described.

4. method as claimed in claim 3, it is characterized in that, when described first page is the sales page of the first commodity, described descriptor is specially: the buyer's guide information of described first commodity or user are to the evaluation information of described first commodity.

5. method as claimed in claim 2, it is characterized in that, be: during the buyer's guide information of at least one commodity described describedly determine to there is with the commodity of described network shop or described network shop the first recommendation information associated by described descriptor in described descriptor, specifically comprise:

6. method as claimed in claim 5, it is characterized in that, described first recommendation information is specially: the keyword of described first network link information and described first network link information; Or the content of pages corresponding to described first network link information; Or the information obtained after the content of pages of described first network link information summarized.

7. method as claimed in claim 5, is characterized in that, describedly carries out search using described shop merchandise classification information as keyword and determines described first recommendation information, is specially:

8. method as claimed in claim 4, is characterized in that, be specially in described descriptor: during the evaluation information of described first commodity, is describedly determined to there is the first recommendation information associated with the commodity of described network shop by described descriptor, specifically comprises:

9. method as claimed in claim 8, it is characterized in that, the described keyword determining the first kind from described first keyword set, specifically comprises:

Determine the type of merchandise information of described first commodity;

10. method as claimed in claim 4, it is characterized in that, be specially in described descriptor: during the buyer's guide information of described first commodity, describedly determined to there is the first recommendation information associated with the commodity of described network shop by described descriptor, specifically comprise:

11. methods as claimed in claim 10, it is characterized in that, described first recommendation information is specially: the keyword of described second network link information and described second network link information; Or the content of pages corresponding to described second network link information; Or the information obtained after the content of pages of described second network link information summarized.

12. methods as claimed in claim 10, is characterized in that, the described brand message using the type of merchandise information of described first commodity and described first commodity obtains described first recommendation information as keyword, specifically comprises:

13. methods as claimed in claim 2, is characterized in that, describedly determine to there is with the commodity of described network shop or described network shop described first recommendation information associated by described descriptor, specifically comprise:

The second keyword set is obtained out from described descriptor;

14. methods as claimed in claim 13, is characterized in that, when described the first keyword in described second keyword set is positioned at the first predetermined keyword list, determines described first recommendation information comprising fisrt feature information, specifically comprise:

15. as arbitrary in claim 2-14 as described in method, it is characterized in that, described method also comprises:

16. methods as claimed in claim 15, it is characterized in that, described negative Transaction Information specifically comprises: the commodity that described network shop is sold are by the quantity of reimbursement, the quantity that the commodity sold of described network shop are complained, the quantity of the commodity transaction failure of described network shop, there is the quantity of dispute in described network shop and buyer, the quantity of the unfavorable ratings information received by described network shop, the quantity of the anonymous ratings that described network shop receives, the quantity of the anonymous deal of described network shop, at least one information in the quantity of the wash sale of described network shop.

17. methods as claimed in claim 15, is characterized in that, described second recommendation information comprises M bar and recommends sub-information, M is positive integer, every bar recommends the negative Transaction Information of the corresponding class of sub-information, in described acquisition and when exporting the second recommendation information, described method also comprises:

18. 1 kinds of signal conditioning packages, is characterized in that, comprising:

19. devices as claimed in claim 18, it is characterized in that, the webpage of described preset kind is specially: shopping class webpage; Described acquisition module, specifically for:

Described determination module, specifically for:

20. devices as claimed in claim 19, it is characterized in that, described descriptor is specially: the buyer's guide information of at least one commodity described; Or user is to the evaluation information of at least one commodity described.

21. devices as claimed in claim 20, it is characterized in that, when described first page is the sales page of the first commodity, described descriptor is specially: the buyer's guide information of described first commodity or user are to the evaluation information of described first commodity.

22. devices as claimed in claim 19, is characterized in that, in described descriptor be: during the buyer's guide information of at least one commodity described, described determination module, specifically comprises:

23. devices as claimed in claim 22, it is characterized in that, described first recommendation information is specially: the keyword of described first network link information and described first network link information; Or the content of pages corresponding to described first network link information; Or the information obtained after the content of pages of described first network link information summarized.

24. devices as claimed in claim 22, is characterized in that, described first search unit, specifically for:

25. devices as claimed in claim 21, is characterized in that, be specially in described descriptor: during the evaluation information of described first commodity, described determination module, specifically comprises:

26. devices as claimed in claim 25, is characterized in that, described 3rd determining unit, specifically comprises:

27. devices as claimed in claim 21, is characterized in that, be specially in described descriptor: during the buyer's guide information of described first commodity, described determination module, specifically comprises:

28. devices as claimed in claim 27, it is characterized in that, described first recommendation information is specially: the keyword of described second network link information and described second network link information; Or the content of pages corresponding to described second network link information; Or the information obtained after the content of pages of described second network link information summarized.

29. devices as claimed in claim 27, is characterized in that, described second search unit, specifically for:

30. devices as claimed in claim 19, it is characterized in that, described determination module, specifically comprises:

31. devices as claimed in claim 30, is characterized in that, described 5th determining unit, specifically comprises:

32. as arbitrary in claim 19-31 as described in device, it is characterized in that, described device also comprises:

33. devices as claimed in claim 32, it is characterized in that, described negative Transaction Information specifically comprises: the commodity that described network shop is sold are by the quantity of reimbursement, the quantity that the commodity sold of described network shop are complained, the quantity of the commodity transaction failure of described network shop, there is the quantity of dispute in described network shop and buyer, the quantity of the unfavorable ratings information received by described network shop, the quantity of the anonymous ratings that described network shop receives, the quantity of the anonymous deal of described network shop, at least one information in the quantity of the wash sale of described network shop.

34. devices as claimed in claim 32, is characterized in that, described second recommendation information comprises M bar and recommends sub-information, and M is positive integer, and every bar recommends the negative Transaction Information of the corresponding class of sub-information, and described device also comprises: