CN109144953B - Search file sorting method, device, equipment, storage medium and search system - Google Patents
Search file sorting method, device, equipment, storage medium and search system Download PDFInfo
- Publication number
- CN109144953B CN109144953B CN201810847629.8A CN201810847629A CN109144953B CN 109144953 B CN109144953 B CN 109144953B CN 201810847629 A CN201810847629 A CN 201810847629A CN 109144953 B CN109144953 B CN 109144953B
- Authority
- CN
- China
- Prior art keywords
- search
- field
- authority
- file
- files
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the application provides a method, a device, equipment, a storage medium and a search system for ranking search files, wherein authority degrees of published objects in all fields are counted in advance, the search field matched with a search request is determined, the authority degrees of the published objects in the search field are further determined according to recalled matched files, and the authority degrees can represent the quality of the matched files, so that the ranking of the matched files is performed based on the authority degrees, and the purpose of ranking according to high quality can be achieved.
Description
Technical Field
The application relates to the technical field of data search, in particular to a method, a device, equipment, a storage medium and a search system for sorting search files.
Background
The search ranking means that when a user inputs keywords for searching, the system matches the recalled related files in the database according to the input keywords, searches and ranks the recalled files according to the matching degree, and then displays the recalled files to the user. It will be appreciated that the higher the quality of the top ranked files, the easier it is to hit the user's search requirements.
For a web page search scenario, the prior art generally ranks web pages according to their PageRank rank, which is the web page quality calculated by a search engine according to the mutual hyperlinks between the web pages, and ranks the web pages according to the web page quality. In other scenes which are not web page search, such as scenes of searching videos and pictures in WeChat public number search articles, microblog search articles and multimedia applications, since searched files lack information similar to hyperlinks in web pages, search sequencing cannot be performed according to the conventional web page PageRank ranking technology.
In view of this, in the prior art, a scheme for sorting search files according to high quality and low quality in a non-web search scenario is needed.
Disclosure of Invention
In view of this, embodiments of the present application provide a sorting method, apparatus, device, storage medium, and search system, which can achieve the purpose of sorting search files according to high quality in a non-web search scenario.
In order to achieve the above purpose, the embodiments of the present application provide the following technical solutions:
in one aspect, the present application provides a method for ranking search files, including:
acquiring matching files recalled based on a search request and a release object corresponding to each matching file;
determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs;
determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field;
and sorting the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain a sorting result of each matched file.
In another aspect, the present application provides a device for sorting search files, including:
the file acquisition unit is used for acquiring matching files recalled based on search requests and release objects corresponding to the matching files;
a search field determining unit, configured to determine a search field that matches the search request, where the search field is a field to which a file that the search request initiator intends to search belongs;
a search field authority degree determining unit, configured to determine authority degrees of the published objects corresponding to each matching file in the search field by referring to authority degrees of preset published objects in each field;
and the ranking adjusting unit is used for ranking the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain the ranking result of each matched file.
In another aspect, the present application further provides a search system, including: a client and a server;
the client is used for receiving a search request initiated by a user and forwarding the search request to the server;
the server is used for acquiring matching files recalled based on search requests and issuing objects corresponding to the matching files; determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs; determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field; sorting the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain a sorting result of each matched file, and returning the sorting result to the client;
and the client is also used for displaying each matched file according to the sorting result.
On the other hand, the application also provides a sorting device for searching files, which comprises: at least one memory and at least one processor; the memory stores programs, and the processor calls the programs stored in the memory to realize the steps of the sorting method for the search files.
In another aspect, the present application also provides a storage medium having a computer program stored thereon, where the computer program is executed by a processor to implement the steps of the method for ranking search files as described above.
Based on the technical scheme, the method for sorting the search files, provided by the embodiment of the application, acquires the matching files recalled based on the search request and the release objects corresponding to the matching files; determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs; determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field; and sequencing the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain the sequencing result of each matched file. Therefore, according to the method and the device, the authority degrees of the issued objects in all fields are counted in advance, the search field matched with the search request is determined, the authority degrees of the issued objects in the search field are determined according to the recalled matched files, the authority degrees can represent the quality of the matched files, and therefore the matched files are sorted based on the authority degrees, and the purpose of sorting according to the quality can be achieved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic diagram illustrating a composition structure of a scene to which the search document ranking method of the present application is applied;
fig. 2 is a signaling interaction diagram of a search file ordering method disclosed in an embodiment of the present application;
FIG. 3 is a diagram illustrating an example of an application of a method for sorting search files according to an embodiment of the present application;
FIG. 4 is a flowchart of a search file ranking method disclosed in the present application from a server-side perspective;
FIG. 5 illustrates a domain distribution histogram;
FIG. 6 is a flowchart of another search file ranking method disclosed in the present application from a server-side perspective;
FIG. 7 illustrates an exemplary diagram of a WeChat search application of an authority determination method;
FIG. 8 is a schematic structural diagram of a search file sorting apparatus disclosed in the present application from a server side perspective;
fig. 9 illustrates a hardware configuration block diagram of the sorting apparatus.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Fig. 1 is a schematic diagram illustrating a composition structure of a scene to which the search document ranking method of the present application is applied.
In the scenario of fig. 1, this includes: a terminal 100 and a server 200.
The terminal 100 may be a mobile electronic device such as a mobile phone and an IPAD, or a non-mobile electronic device such as a computer.
The terminal 100 runs a client, which is a client providing a file search function, such as a WeChat application, a microblog application, a web application, a multimedia application, a picture application, and the like, and the client can be used by a user to input a search request, and obtain a ranked matching file corresponding to the search request, and then display the ranked matching file. The matching file refers to a file which is obtained by searching based on the search request and is matched with the search request. The user can click on the displayed matching files and browse the matching files.
Optional scenes are that a user searches for articles in a WeChat search scene, or searches for articles in a microblog search scene, or searches for video information uploaded by a registered user in a multimedia application search video scene, or searches for picture information uploaded by the registered user in a picture application search picture scene.
Optionally, the terminal 100 may support a client to implement local file search and/or file sorting.
Such as a client that may search locally for matching files in response to a user search request.
For another example, the client may rank the matching files based on the matching files obtained by local search or the matching files returned by the server, and when ranking, first determine the search field matched with the search request, and then determine the authority of the release object in the search field for the matching files, and rank the matching files based on the authority.
Further optionally, file searching and/or file sorting may also be implemented by the server 200.
For example, the client can forward the search request of the user to the server, and the server searches the matching file.
Further, the server may rank the matching files based on matching files searched locally or matching files uploaded by the client. The ranking rule is the same as the foregoing rule, that is, the search field matched with the search request is determined first, the authority of the issued object in the search field is further determined for the matching file, and the matching file is ranked based on the authority.
Generally, for the sorted matching files, they can be presented to the user through the terminal 100. Of course, the ranking result of each matching file may also be stored by the terminal 100 or the server 200, and passed to other modules or externally called.
The server 200 may be a server corresponding to a client running on the terminal 100, and may be a search server, a query server, or other specially configured service device. Taking the client as an example of the wechat application, the server 200 may be a wechat server.
Next, with reference to a signaling interaction diagram illustrated in fig. 2, an implementation manner of a method for sorting search files in an embodiment of the present application is described, in the implementation manner of the present embodiment, a file search process and a file sorting process are both implemented at a server 200, and a file display is shown as an example at a terminal 100, as shown in fig. 2, the method includes:
step S10, the terminal receives a search request initiated by the user.
Specifically, the client interface of the terminal may provide a search interface through which a user may input a search request. The search request typically includes a search term.
For example, a user wants to search for a science and technology assessment article related to a mechanical keyboard under a WeChat search scene. A "mechanical keypad" may be entered as a search request in a search input box provided by the wechat-search interface.
And step S11, the terminal forwards the search request to the server.
Step S12, the server obtains the matching files recalled based on the search request, and the publishing object corresponding to each matching file.
Specifically, after receiving the search request, the server may search for a matching file in the database based on the search term included in the search request, and then recall the matching file, where the matching file is a file matching the search request, and more specifically, may be a file matching the search term included in the search request. On the basis, the server further obtains the release object corresponding to the matching file. Further, the server can also obtain the field to which the matching file belongs.
And the publishing object corresponding to the matching file is the object for publishing the matching file. The release object can be in the form of an account number, such as a wechat public number, a microblog account number and the like.
The method and the device count the files issued by each issuing object in advance, and gather the classification of the fields to which the files belong, such as N types of fields. Each published file is labeled with the domain to which it belongs. For example, when a publishing object publishes a file, the domain affiliation of the published file needs to be selected, or the domain of the published file needs to be labeled by itself. In addition, the method and the device can determine the field of the matching file in a text analysis mode according to the content of the matching file.
And step S13, the server determines the search domain matched with the search request.
The search field is a field to which a file intended to be searched by a search request initiator belongs. The present embodiment may determine the search area matching the search request in various ways. Such as determining a search area based on the content contained in the search request, or determining a search area based on recalled matching files, etc.
And step S14, the server side determines the authority of the issued object corresponding to each matching file in the search field by referring to the authority of the preset issued object in each field.
In this embodiment, the authority of each published object in each field is obtained in advance. The authority of the published object i in the jth field can be understood as the degree of general recognition of the file in the jth field published by the published object i by the jth field. The authority degree can represent the quality of the file in the jth field issued by the issuing object i, and the higher the authority degree is, the higher the representative quality is.
For example, the account number is 3000286699, and the issue is named as "clove mom", and the issue relates to three fields, namely "child care", "health", and "pet". The authority score of the user in the 'child bearing' field is very high and can reach 1, the authority score of the user in the 'health' field is high and can reach 0.712043, and the authority score of the user in the 'pet' field is general and can reach 0.357508. In other areas, the user has zero acceptance of their published document, and thus the authority score is 0.
The authority of each published object in each domain may be stored in a list form, or in other forms. See table 1 below to illustrate one storage mode:
TABLE 1
And determining the authority degree of the published object corresponding to each matching file in the search field based on the authority degrees of the published objects in the fields obtained in advance.
Still taking the example in table 1 as an example, assuming that the search field is field 2, and the corresponding issued object of the matching file is account 1, the authority of account 1 in field 2 is S12.
And step S15, the server ranks the matching files according to the authority of the issued object corresponding to each matching file in the search field to obtain the ranking result of each matching file.
Specifically, in the step S12, the server obtains the matching files recalled based on the search request, which may be preliminarily sorted according to the matching degree, and on this basis, in this step, the matching files are sorted according to the authority of the corresponding release object of each matching file in the search field. In another case, the matching files obtained in step S12 are not initially sorted according to the matching degree, and the sorting process in this step may be understood as sorting the matching files according to the authority of the published object corresponding to each matching file in the search field.
It can be understood that the higher the authority of the corresponding published object in the search field, the higher the quality of the matching file is characterized, and the file which the user intends to search is more likely to be, so the ranking position of the matching file should be the higher.
And step S16, the server returns the sequencing result of each matching file to the terminal.
And step S17, the terminal displays the matched files according to the sorting result.
According to the method for ranking the search files provided by the embodiment of the application, the authority degrees of all published objects in all fields are stored in the server side in advance, the search fields matched with the search requests are determined, the authority degrees of the published objects in the search fields are determined according to the recalled matched files, the authority degrees can represent the quality of the matched files, therefore, the matched files are ranked based on the authority degrees, the purpose of ranking according to the quality can be achieved, the ranking result of the matched files is returned to the terminal, the matched files are displayed by the terminal according to the sequence, a user can obtain the files which the user intends to search through the files displayed in front, the click conversion rate is improved, and the search experience of the user is improved.
With reference to fig. 3, an application example of the method for ranking search files provided in the embodiment of the present application may be as follows:
when searching for a document, a user can input a search word in a search box of a search interface of a terminal (such as a notebook computer, a mobile phone and the like), the user clicks a search button of the search box, and the terminal can generate a search request according to the search word input by the user and submit the search request to a server.
The server can store the authority of each object in each field in advance. And searching the matched document and the release object corresponding to the matched document in the database according to the search request sent by the terminal. The server determines the search field matched with the search request and determines the authority degree of the issued object corresponding to each matched document in the search field. Further, according to the authority degree of the issuing object corresponding to each matched document in the search field, all the matched documents are ranked to obtain the ranking result of all the matched documents.
The server returns the sorting result to the terminal, and the terminal displays the matching documents on the search display interface, where the matching documents listed in the top three in the example of fig. 3 are: document 1, document 2, and document 3.
Of course, besides the application of document searching, the embodiment of the present application can also be applied in other file searching scenarios, and the specific application mode can be adjusted according to the actual situation.
Next, the embodiment of the present application introduces a method for ranking search files of the present application from the server side.
As shown in fig. 4, the method may include:
step S100, obtaining the matching files recalled based on the search request and the release objects corresponding to the matching files.
Specifically, the server may search for a matching file in the database based on the search request, and then recall the matching file. On the basis, the server further obtains the release object corresponding to the matching file. Further, the server can also obtain the field to which the matching file belongs.
And the publishing object corresponding to the matching file is the object for publishing the matching file. The release object can be in the form of an account number, such as a wechat public number, a microblog account number and the like.
The method and the device count the files issued by each issuing object in advance, and gather the classification of the fields to which the files belong, such as N types of fields. Each published file is labeled with the domain to which it belongs. For example, when a publishing object publishes a file, the domain affiliation of the published file needs to be selected, or the domain of the published file needs to be labeled by itself. In addition, the method and the device can determine the field of the matching file in a text analysis mode according to the content of the matching file.
And step S110, determining a search field matched with the search request.
Wherein the search field is a field to which a file intended to be searched by the search request initiator belongs.
The present embodiment may determine the search area matching the search request in various ways. Such as determining a search area based on the content contained in the search request, or determining a search area based on a recalled matching file, etc.
And step S120, referring to the authority degrees of preset issued objects in all fields, and determining the authority degree of the issued object corresponding to each matched file in the search field.
In this embodiment, the authority of each published object in each field is obtained in advance. The authority of the published object i in the jth field can be understood as the degree of general recognition of the file in the jth field published by the published object i by the jth field. The authority degree can represent the quality of the file in the jth field issued by the issuing object i, and the higher the authority degree is, the higher the representative quality is.
And determining the authority degree of the published object corresponding to each matching file in the search field based on the authority degrees of the published objects in the fields obtained in advance.
Step S130, ranking each matching file according to the authority degree of the issued object corresponding to each matching file in the search field to obtain the ranking result of each matching file.
It can be understood that the higher the authority of the corresponding published object in the search field, the higher the quality of the matching file is characterized, and the file which the user intends to search is more likely to be, so the ranking position of the matching file should be the higher.
In the specific implementation process, the ranking weight of the matched files can be determined according to the authority degrees of the published objects corresponding to the matched files in the search field, and the higher the authority degree is, the larger the corresponding ranking weight is.
Further, the ranking weight of each matching file is referred to, and ranking adjustment is performed on each matching file based on a rule that the greater the ranking weight is, the more the corresponding ranking position is, the higher the ranking position is.
According to the ranking method for the search files, the authority degrees of all issued objects in all fields are stored in the server side in advance, the search fields matched with the search requests are determined, the authority degrees of the issued objects in the search fields are determined according to the recalled matched files, the authority degrees can represent the quality of the matched files, and therefore the ranking of all the matched files based on the authority degrees can achieve the purpose of ranking according to high quality.
Furthermore, in the embodiment, the recalled matched files are sorted and adjusted according to the authority of the published objects in the search field, so that the problem that when the search field is the current hotspot field, some published objects in non-search fields rub against the files in the hotspot publishing search field, the files belong to the search field but are limited to the self-capability of the published objects, the published files are low in quality, and the published files are possibly sorted forward according to the prior art, so that the search experience of a user is influenced is solved.
In another embodiment of the present application, a process of determining a search area matching the search request in step S110 is described.
This embodiment introduces two implementation manners, which are as follows:
first, a search domain is determined based on the recalled matching files. The details are as follows:
1) and counting the distribution condition of the field to which each matching file belongs.
Specifically, each matching file has a corresponding domain, and the distribution condition of the domains can be counted based on the recalled matching files.
Referring to fig. 5, a domain distribution histogram is illustrated.
In the case illustrated in fig. 5, by counting the domain affiliations of the matching files, three domains are found to be affiliated to the matching files, which are: education, health, science and technology. The distribution of each domain is shown in fig. 5, and the distribution rates are respectively: health-education-science and technology.
It should be noted that the distribution of the domains can also be embodied in other forms, and fig. 5 only illustrates one case shown by a histogram.
2) And determining the field with the highest distribution rate as the search field matched with the search request according to the distribution condition.
Still taking the case illustrated in fig. 5 as an example, the area with the highest distribution rate is technology, and thus the search area matched with the search request can be determined to be technology.
In the method for determining the search domain in this embodiment, starting from the perspective of the matching file recalled based on the search request, the domain with the highest distribution rate is selected as the search domain by analyzing the distribution of the home domain of the matching file.
Next, another way of determining the search area is described.
Second, a search domain is determined based on the search request. The details are as follows:
1) a matching search intent is determined based on the search request.
In particular, the search intent indicates the complete information that the user intends to search through the search request. The complete information not only contains the object which is intended to be searched, but also comprises other attributes of the object, such as commodity attributes, technology attributes and the like.
For example, a user search request includes the search term "mechanical keyboard". The real search intention of the user is "want to refer to a science and technology evaluation article about the mechanical keyboard". The simple search term "mechanical keyboard" may also correspond to other search intentions, such as "selling price of mechanical keyboard" and the like.
In this step, a matching search intention is determined based on the search request, where the matching search intention is a search intention that is actually to be expressed by the initiator of the search request.
In the specific implementation process, the search intention can be accurately determined by combining the personal portrait data of the search request initiator.
2) And determining a domain corresponding to the search intention as a search domain matched with the search request.
Specifically, different search intents may correspond to different domains, and the domain corresponding to the search intention is determined in this step as the search domain matched with the search request.
In the specific implementation process, corresponding relations between different keywords and the fields can be constructed in advance, and then the corresponding fields are determined as the fields corresponding to the search intents according to the keywords contained in the search intents. For example, the keyword "science and technology evaluation" corresponds to a field of "science and technology", and the keyword "body" and the keyword "medicine" correspond to a field of "health", and the like.
In the method for determining the search field in this embodiment, starting from the perspective of the search request, the search request is analyzed to determine the matching search intention, and then the corresponding field is determined according to the search intention as the search field.
In another embodiment of the present application, a method for ranking search documents of the present application is further explained. As shown in fig. 6, the sorting method may include:
step S200, obtaining the matching files recalled based on the search request and the release objects corresponding to the matching files.
And step S210, determining a search field matched with the search request.
Specifically, steps S200 to S210 in this embodiment correspond to steps S100 to S110 in the foregoing embodiment one to one, and refer to the foregoing description in detail, which is not repeated herein.
Step S220, according to the field to which each matching file belongs and the field similarity of the search field, determining the preliminary ranking of each matching file.
Specifically, each matching file exists in the corresponding field, and in this step, each matching file may be initially sorted according to the field to which the matching file belongs and the sequence of high-to-low similarity of the field to the search field.
It is understood that the higher the domain similarity between the domain to which the matching document belongs and the search domain, the higher the preliminary ranking of the corresponding matching document.
Optionally, the similarity between different domains may be pre-labeled, such as by text similarity algorithm calculation or manual calibration. Based on the similarity, the similarity between the field of each matching file and the search field can be determined, and then the matching files are initially sorted according to the similarity.
Optionally, there is no strict order restriction between step S220 and step S230, and the steps may be executed in any order or synchronously.
Step S230, referring to the authority degrees of preset published objects in each field, and determining the authority degree of the published object corresponding to each matching file in the search field.
Step S240, according to the authority of the issued object corresponding to each matching file in the search field, performing ranking adjustment on each matching file to obtain a ranking result of each matching file.
Specifically, in the step S220, the preliminary ranking of each matching file is determined, in this step, the preliminary ranking result of the matching file is further ranked and adjusted according to the authority of the corresponding publishing object of the matching file in the search field, wherein the ranking position with a high authority can be adjusted forward, the ranking position with a low authority can be moved backward, and the ranking result of the matching file is obtained after the adjustment.
The method for sorting search files provided in this embodiment further performs sorting adjustment on each matching file by combining authority of a distribution object corresponding to the matching file in the search field on the basis of performing preliminary sorting according to similarity between the field to which the matching file belongs and the search field, so that a sorting result is more reasonable.
In yet another embodiment of the present application, a process for determining the authority of a published object in various domains is presented.
The authority degree statistics can be performed according to a set period, for example, every other statistical time period, and the authority degrees of the issued objects in each field are determined according to the search logs in the statistical time period. Wherein, the statistical time period may be one day, one month or other time period.
The method and the device for determining the initial authority according to the click characteristics of the user can specifically include the following steps:
1) and determining N target issuing objects corresponding to the clicked file in the statistical time period.
Specifically, according to the search log, files clicked by the user within a statistical time period can be determined, and then target publishing objects corresponding to the clicked files are determined, wherein N determined target publishing objects are defined.
2) And determining the initial authority of the ith target issuing object in the jth field according to the times of the file issued by the ith target issuing object, recalled by the search request of the jth field and clicked.
Here, the value of i [1, N ], the value of j [1, M ], M is the total number of fields.
For the ith target publishing object, when the initial authority of the ith target publishing object in the jth field is determined, the number of times that the file published by the ith target publishing object is recalled and clicked by the search request of the jth field can be combined.
The corresponding initial authority calculation method may be as follows:
Scoreij＝∑log(Xij) Equation 1
Wherein, ScoreijAnd the initial authority score of the ith target release object in the jth field is shown. XijAnd the number of times that the file published by the ith target publishing object is recalled and clicked by the search request of the jth field is represented, wherein the search request of the jth field can be understood as the search request corresponding to the search intention belonging to the jth field.
Further, considering that the clicked file in the aforementioned 1) may not be recalled corresponding to the search request based on the strong search intention, such number of clicks may have a negative influence on the authority determination result, the aforementioned 1) may be implemented as follows:
and determining N target release objects corresponding to the files recalled by the search requests and clicked by the strong search intentions within the statistical time period.
Wherein the strong search intention is a search intention having a search intention score exceeding a preset intention score.
On this basis, the above 2) can be realized as follows:
and determining the initial authority of the ith target issuing object in the jth field according to the file issued by the ith target issuing object, the number of times that the search request corresponding to the strong search intention of the jth field is recalled and clicked, and the score of the strong search intention.
The corresponding initial authority calculation method may be as follows:
Scoreij＝∑(log(Xij)*Qj) Equation 2
Wherein, ScoreijAnd the initial authority score of the ith target release object in the jth field is shown. XijThe number of times Q that the document which represents the ith target publishing object is recalled and clicked by the search request corresponding to the strong search intention of the jth fieldjA score representing a strong search intention for the jth domain. The object of the formula sum is a strong search intention per j domain.
Further, on the basis of calculating the initial authority of the above example, a process of performing error adjustment on the initial authority according to click characteristics of the user corresponding to the popular search intention may be added.
The popular search intentions refer to topY search intentions with the highest heat counted in the last statistical time period.
The specific adjustment process may include:
determining a first authority adjustment value of the ith target published object in the jth domain in conjunction with one or more of the following exemplary items, wherein the items may include:
the file published by the ith target publishing object is recalled and clicked by the search request corresponding to the popular search intention of the jth field, the ratio of the number of times (specifically, the number of times recalled and clicked by the search request corresponding to the popular search intention of the jth field) of the number of times to the total number of times of clicked files published by the ith target publishing object, and the score of the popular search intention.
Specifically, the first authority adjustment value may be determined according to the above formula 1 or formula 2, and only X in the above formula needs to be determinedijRepresenting the number of times that the file published as the ith target publication object is recalled and clicked by the search request corresponding to the hit search intention of the jth field, or the ratio of the number of times to the total number of times that the file published by the ith target publication object is clicked, and QjExpressed as the score of the hit search intention of the jth field.
After the first authority adjustment value is determined, the initial authority of the ith target issuing object in the jth field and the first authority adjustment value may be added to obtain the adjusted initial authority of the ith target issuing object in the jth field.
Further, on the basis of the above embodiment, a process of adjusting the initial authority or the adjusted initial authority again according to the historical text information of the issuing object may be added.
First, a process of determining an initial authority adjustment from the perspective of a text similarity is introduced, which specifically includes:
1) determining the similarity between the file issued by the ith target issuing object and the file issued by the set number (such as topL) issuing objects with the highest initial authority in the jth field in the statistical time period, and taking the similarity as the similarity between the ith target issuing object and the jth field.
Specifically, the files can be vectorized through the doc2vec model, and then the similarity between the vectorized files is calculated.
Referring to table 2 below, a search-for-first scene in WeChat public numbers is taken as an example, which shows the initial authority scores of several public numbers with the highest initial authority in the field of chicken soup.
TABLE 2
2) And determining a second authority degree adjusting value of the ith target issuing object in the jth field according to the similarity of the ith target issuing object and the jth field.
It can be understood that, the higher the similarity between the ith target issuing object and the jth field is, the higher the quality of the file of the jth field indicating that the ith target issuing object issues is, the higher the authority of the corresponding ith target issuing object in the jth field should be, so that the second authority adjustment value belongs to a positive number and increases as the similarity increases.
3) And adding the initial authority of the ith target issuing object in the jth field and the second authority adjustment value to obtain the adjusted authority.
Further, the process of determining the initial authority adjustment from the perspective of the complexity of the issue text category may specifically include:
1) and determining a third authority degree adjusting value of the ith target issuing object in each field according to the category complexity of the field to which the ith target issuing object issuing file belongs in the statistical time period.
It can be understood that, the higher the complexity of the kind of the field to which the publishing file belongs, the more messy the text of the ith target publishing object is, and the field is not specific enough, so that the authority of the ith target publishing object in each field needs to be punished, and the third authority can be set to a smaller value, for example, to a negative number.
2) And adding the initial authority of the ith target issuing object in each field and the third authority adjustment value to obtain the adjusted authority.
Still further, the process of determining the initial authority adjustment from the perspective of the proportion of the text field and the influence of the text field may specifically include:
determining a fourth authority adjustment value of the ith target published object in the jth domain in conjunction with one or more of the following exemplary items, wherein the items may include:
in the statistical time period, the ratio of the number of files belonging to the jth field in the files issued by the ith target issuing object to the total number of the files issued by the ith target issuing object, the ratio of the number of files belonging to the jth field and having a clicked number exceeding a preset number threshold value in the files issued by the ith target issuing object to the total number of the files issued by the ith target issuing object, and the influence of the files issued by the ith target issuing object and belonging to the jth field are counted.
It can be understood that, the larger the ratio of the number of files belonging to the jth field to the total number of files issued by the ith target issuing object is, the larger the fourth authority adjustment value of the ith target issuing object in the jth field is.
The files which belong to the jth field and of which the clicked number exceeds the preset number threshold value can be understood as high click rate files of the jth field. It can be understood that, the larger the ratio of the number of the high-click-count files to the total number of the files published by the ith target publishing object is, the larger the fourth authority adjustment value of the ith target publishing object in the jth field is.
The influence of the file belonging to the jth field and published by the ith target publishing object can be determined by parameters such as the number of clicks, the number of forwarding, the number of praise, the number of comments and the like. It can be understood that, the larger the influence of the file belonging to the jth domain published by the ith target publishing object is, the larger the fourth authority adjustment value of the ith target publishing object in the jth domain is.
Based on the fourth authority adjustment value of the ith target issuing object in the jth field, the fourth authority adjustment value may be added to the initial authority value of the ith target issuing object in the jth field to obtain the adjusted authority.
Furthermore, on the basis of the above embodiment, a process of adjusting the initial authority or the adjusted initial authority again according to the static quality of the published object and the friend relationship may be added.
The specific implementation process may include:
1) and determining a fifth authority adjustment value of the ith target release object in each field according to the static quality of the ith target release object and the friend relationship of other release objects.
The static quality of the ith target issuing object can be determined according to parameters such as the completeness of data filling, the number of issued documents, the number of forwarded and approved documents, the registration time, the number of average issued documents per day, the maximum number of continuous issued documents and the like.
The friend relationship between the ith target release object and other release objects may be the number of fans of the ith target release object, and the value is in direct proportion to the fifth authority adjustment value.
2) And adding the initial authority of the ith target issuing object in each field and the fifth authority adjustment value to obtain the adjusted authority.
To sum up, the embodiment of the present application discloses a determination process of an initial authority value of an ith target issuing object in a jth field and a determination process of first to fifth authority adjustment values.
Fig. 7 illustrates an overall scheme of authority comprehensive scoring, taking a public number first search scene as an example. From the above, it can be divided into three directions, namely: the method comprises the steps of determining initial authority of the public number based on user click characteristics, determining an authority adjustment value based on historical public number text sending information, and determining the authority adjustment value based on public number quality characteristics.
The process of determining the initial authority of the public account based on the user click features can be divided into the following steps: determining initial authority based on the number of times that the public number issue is recalled by any field request and clicked; and adjusting the initial authority error based on the number of strong search intention clicks.
The process of determining the authority adjustment value based on the historical public number issue information can be divided into the following steps: determining an authority degree adjusting value based on the similarity between the historical published articles of the public number and the published articles of the head public number, and determining the authority degree adjusting value based on the spectroscope result of the historical published articles of the public number. The historical textbook spectroscopic results may include, among others: the method comprises the following steps of sending a sentence type complexity, a sending sentence occupation ratio in any field, a high reading article occupation ratio published in any field and the influence of publishing articles in any field.
The process of determining the authority adjustment value based on the public account quality characteristics may include: static quality of public and number of vermicelli.
Referring to table 3, a first search scenario in the WeChat public number is taken as an example, which shows authority values of different domains of a part of the public number.
TABLE 3
Taking the public number "free-running hornet cell" as an example, according to the above table 3, the authority value in the field of "Travel" is 0.834340, the authority value in the field of "Food" is 0.201668, and the authority value in the field of "App" is 0.21520.
In table 3 above, for any public number, if the authority value in a certain field is not marked, it indicates that the authority value in the field is 0.
The following introduces a search file sorting apparatus provided in the embodiment of the present application, and the search file sorting apparatus described below may be considered as a program module that is required to be set by a server to implement the search file sorting apparatus provided in the embodiment of the present application. The contents of the search file ranking device described below may be referred to in correspondence with the contents of the search file ranking method described above.
Fig. 8 is a block diagram of a structure of a search file ranking apparatus according to an embodiment of the present application, where the apparatus is applicable to a server, and referring to fig. 8, the search file ranking apparatus may include:
a file obtaining unit 11, configured to obtain matching files recalled based on search requests, and a release object corresponding to each matching file;
a search domain determining unit 12 configured to determine a search domain matching the search request, where the search domain is a domain to which a file intended to be searched by the search request initiator belongs;
a search domain authority degree determining unit 13, configured to determine authority degrees of the published objects corresponding to each matching file in the search domain by referring to authority degrees of preset published objects in each domain;
and the sorting unit 14 is configured to sort the matching files according to the authority of the issued object corresponding to each matching file in the search field, so as to obtain a sorting result of each matching file.
Optionally, the embodiment of the present application illustrates two optional structures of the search area determining unit, which are respectively as follows:
the first search area determination unit may include:
the distribution condition counting unit is used for counting the distribution condition of the field to which each matched file belongs;
and the distribution rate determining unit is used for determining the field with the highest distribution rate as the search field matched with the search request according to the distribution condition.
The second search area determination unit may include:
a search intention determining unit for determining a matching search intention based on the search request;
and the intention field corresponding unit is used for determining the field corresponding to the search intention as the search field matched with the search request.
Optionally, the apparatus of the present application may further include:
and the preliminary ranking unit is used for determining the preliminary ranking of each matching file according to the domain similarity of the domain to which each matching file belongs and the search domain before ranking and adjusting each matching file.
Optionally, the preliminary sorting unit may include:
and the similarity sorting unit is used for preliminarily sorting the matched files according to the sequence that the similarity between the field to which the matched files belong and the field of the search field is from high to low.
Optionally, the sorting adjusting unit may include:
the ranking weight determining unit is used for determining the ranking weight of the matched file according to the authority degree of the release object corresponding to the matched file in the search field, wherein the higher the authority degree is, the larger the corresponding ranking weight is;
and the ranking weight using unit is used for referring to the ranking weight of each matched file and performing ranking adjustment on each matched file based on a rule that the greater the ranking weight is, the more front the corresponding ranking position is.
Optionally, the apparatus of the present application may further include: an authority determination unit, configured to determine authority of each published object in each field, where the authority determination unit may include:
the target issuing object determining unit is used for determining N target issuing objects corresponding to the clicked files in the statistical time period;
and the initial authority determining unit is used for determining the initial authority of the ith target issuing object in the jth field according to the times that the document issued by the ith target issuing object is recalled by the search request in the jth field and clicked, wherein i takes a value of [1, N ], j takes a value of [1, M ], and M is the total number of the fields.
Optionally, the target published object determining unit may include:
the first target issuing object determining subunit is used for determining N target issuing objects corresponding to the files which are recalled by the search requests corresponding to the strong search intentions and clicked within the statistical time period; the strong search intention is a search intention of which the search intention score exceeds a preset intention score;
correspondingly, the initial authority determination unit may include:
and the first initial authority determining subunit is used for determining the initial authority of the ith target issuing object in the jth field according to the file issued by the ith target issuing object, the number of times that the search request is recalled and clicked by the strong search intention corresponding to the jth field, and the score of the strong search intention.
Optionally, the authority determination unit may further include:
the first adjusting unit is used for determining a first authority adjusting value of the ith target issuing object in the jth field according to the number of times that the file issued by the ith target issuing object is recalled and clicked by the search request corresponding to the popular search intention of the jth field, the ratio of the number of times to the total number of times that the file issued by the ith target issuing object is clicked, and/or the score of the popular search intention; and adding the initial authority of the ith target issuing object in the jth field and the first authority adjustment value to obtain the adjusted authority.
Optionally, the authority determination unit may further include:
the second adjusting unit is used for determining the similarity between the file issued by the ith target issuing object and the files issued by the set number of issuing objects with the highest initial authority in the jth field in the statistical time period, and taking the similarity as the similarity between the ith target issuing object and the jth field; determining a second authority degree adjusting value of the ith target issuing object in the jth field according to the similarity of the ith target issuing object and the jth field; and adding the initial authority of the ith target issuing object in the jth field and the second authority adjustment value to obtain the adjusted authority.
Optionally, the authority determination unit may further include:
a third adjusting unit, configured to determine, according to the class complexity of the domain to which the ith target publishing object publishing file belongs within the statistical time period, a third authority adjustment value of the ith target publishing object in each domain; the higher the category complexity is, the smaller the third authority adjustment value is; and adding the initial authority of the ith target issuing object in each field and the third authority adjustment value to obtain the adjusted authority.
Optionally, the authority determination unit may further include:
a fourth adjusting unit, configured to determine a fourth authority adjustment value of the ith target publishing object in the jth field according to a ratio of the number of files belonging to the jth field in files published by the ith target publishing object to the total number of files published by the ith target publishing object, a ratio of the number of files belonging to the jth field and having a clicked number exceeding a preset number threshold in files published by the ith target publishing object to the total number of files published by the ith target publishing object, and/or an influence of the files belonging to the jth field published by the ith target publishing object in the jth field in a statistical time period; and adding the initial authority of the ith target issuing object in the jth field and the fourth authority adjustment value to obtain the adjusted authority.
Optionally, the authority determination unit may further include:
a fifth adjusting unit, configured to determine a fifth authority adjustment value of the ith target publishing object in each field according to the static quality of the ith target publishing object and the friend relationship with other publishing objects; and adding the initial authority of the ith target issuing object in each field and the fifth authority adjustment value to obtain the adjusted authority.
The sorting device for the search files provided by the embodiment of the application can be applied to sorting equipment for the search files, such as a server and the like; alternatively, fig. 9 shows a block diagram of a hardware structure of the sorting apparatus, and referring to fig. 9, the hardware structure of the sorting apparatus may include: at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;
in the embodiment of the application, the number of the processor 1, the communication interface 2, the memory 3 and the communication bus 4 is at least one, and the processor 1, the communication interface 2 and the memory 3 complete mutual communication through the communication bus 4;
the processor 1 may be a central processing unit CPU or an ASIC specific integrated circuit
(Application Specific Integrated Circuit), or one or more Integrated circuits or the like configured to implement embodiments of the present Application;
the memory 3 may include a high-speed RAM memory, and may further include a non-volatile memory (non-volatile memory) or the like, such as at least one disk memory;
wherein the memory stores a program and the processor can call the program stored in the memory, the program for:
acquiring matching files recalled based on a search request and a release object corresponding to each matching file;
determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs;
determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field;
and sorting the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain a sorting result of each matched file.
Alternatively, the detailed function and the extended function of the program may be as described above.
Embodiments of the present application further provide a storage medium, where a program suitable for execution by a processor may be stored, where the program is configured to:
acquiring matching files recalled based on a search request and a release object corresponding to each matching file;
determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs;
determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field;
and sorting the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain a sorting result of each matched file.
Alternatively, the detailed function and the extended function of the program may be as described above.
An embodiment of the present application further provides a search system, including: a client and a server;
the client is used for receiving a search request initiated by a user and forwarding the search request to the server;
the server is used for acquiring the matching files recalled based on the search request, the field to which each matching file belongs and the corresponding release object; determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs; determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field; according to the authority degree of the issued object corresponding to each matching file in the search field, carrying out sequencing adjustment on each matching file to obtain a sequencing result of each matching file, and returning the sequencing result to the client;
and the client is also used for displaying each matched file according to the sorting result.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the disclosure. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (14)
1. A method of ranking search files, comprising:
acquiring matching files recalled based on a search request and a release object corresponding to each matching file;
determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs;
determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field;
sorting the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain a sorting result of each matched file;
the process for determining the authority of the issued object in each field comprises the following steps:
determining N target release objects corresponding to files which are recalled by the search requests and clicked corresponding to the strong search intentions within the statistical time period; the strong search intention is a search intention of which the search intention score exceeds a preset intention score;
according to the file issued by the ith target issuing object, the number of times that the strong search intention of the jth field corresponds to the search request and is recalled and clicked and the score of the strong search intention, determining the initial authority of the ith target issuing object in the jth field, wherein i takes a value of [1, N ], j takes a value of [1, M ], and M is the total number of the fields.
2. The ranking method of claim 1, wherein the determining a search area matching the search request comprises:
counting the distribution condition of the field to which each matching file belongs;
and determining the field with the highest distribution rate as the search field matched with the search request according to the distribution condition.
3. The ranking method of claim 1, wherein the determining a search area matching the search request comprises:
determining a matching search intent based on the search request;
and determining a domain corresponding to the search intention as a search domain matched with the search request.
4. The ranking method of claim 1, wherein prior to said ranking each of said matching files, the method further comprises:
and determining the preliminary ranking of each matching file according to the domain to which each matching file belongs and the domain similarity of the search domain.
5. The ranking method according to claim 1, wherein the ranking the matching documents according to the authority of the published object corresponding to each matching document in the search field includes:
determining the ranking weight of the matched file according to the authority degree of the issued object corresponding to the matched file in the search field, wherein the higher the authority degree is, the larger the corresponding ranking weight is;
and referring to the ranking weight of each matched file, and ranking each matched file based on a rule that the greater the ranking weight is, the more front the corresponding ranking position is.
6. The sorting method according to claim 1, further comprising:
determining a first authority degree adjusting value of the ith target issuing object in the jth field according to the number of times that the file issued by the ith target issuing object is recalled and clicked by the search request corresponding to the popular search intention of the jth field, the ratio of the number of times to the total number of times that the file issued by the ith target issuing object is clicked, and/or the score of the popular search intention;
and adding the initial authority of the ith target issuing object in the jth field and the first authority adjustment value to obtain the adjusted authority.
7. The sorting method according to claim 1, further comprising:
determining the similarity between the files issued by the ith target issuing object and the files issued by the set number of issuing objects with the highest initial authority in the jth field in the statistical time period as the similarity between the ith target issuing object and the jth field;
determining a second authority degree adjusting value of the ith target issuing object in the jth field according to the similarity of the ith target issuing object and the jth field;
and adding the initial authority of the ith target issuing object in the jth field and the second authority adjustment value to obtain the adjusted authority.
8. The sorting method according to claim 1, further comprising:
determining a third authority degree adjusting value of the ith target issuing object in each field according to the category complexity of the field to which the ith target issuing object issuing file belongs in the statistical time period; the higher the category complexity is, the smaller the third authority adjustment value is;
and adding the initial authority of the ith target issuing object in each field and the third authority adjustment value to obtain the adjusted authority.
9. The sorting method according to claim 1, further comprising:
determining a fourth authority adjustment value of the ith target issuing object in the jth field according to the ratio of the number of files belonging to the jth field in the files issued by the ith target issuing object to the total number of files issued by the ith target issuing object, the ratio of the number of files belonging to the jth field and having a clicked number exceeding a preset number threshold value in the files issued by the ith target issuing object to the total number of files issued by the ith target issuing object, and/or the influence of the files belonging to the jth field issued by the ith target issuing object in the jth field in the statistical time period;
and adding the initial authority of the ith target issuing object in the jth field and the fourth authority adjustment value to obtain the adjusted authority.
10. The sorting method according to claim 1, further comprising:
determining a fifth authority adjustment value of the ith target release object in each field according to the static quality of the ith target release object and the friend relationship with other release objects;
and adding the initial authority of the ith target issuing object in each field and the fifth authority adjustment value to obtain the adjusted authority.
11. A ranking apparatus for searching for a file, comprising:
the file acquisition unit is used for acquiring matching files recalled based on search requests and release objects corresponding to the matching files;
a search field determining unit, configured to determine a search field that matches the search request, where the search field is a field to which a file that the search request initiator intends to search belongs;
a search field authority degree determining unit, configured to determine authority degrees of the published objects corresponding to each matching file in the search field by referring to authority degrees of preset published objects in each field;
the sorting unit is used for sorting the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain a sorting result of each matched file;
the process for determining the authority of the issued object in each field comprises the following steps:
determining N target release objects corresponding to files which are recalled by the search requests and clicked corresponding to the strong search intentions within the statistical time period; the strong search intention is a search intention of which the search intention score exceeds a preset intention score;
according to the file issued by the ith target issuing object, the number of times that the strong search intention of the jth field corresponds to the search request and is recalled and clicked and the score of the strong search intention, determining the initial authority of the ith target issuing object in the jth field, wherein i takes a value of [1, N ], j takes a value of [1, M ], and M is the total number of the fields.
12. A search system, comprising: a client and a server;
the client is used for receiving a search request initiated by a user and forwarding the search request to the server;
the server is used for acquiring matching files recalled based on search requests and issuing objects corresponding to the matching files; determining a search field matched with the search request, wherein the search field is a field to which a file intended to be searched by a search request initiator belongs; determining authority degrees of the issued objects corresponding to each matching file in the search field by referring to the authority degrees of preset issued objects in each field; sorting the matched files according to the authority degree of the issued object corresponding to each matched file in the search field to obtain a sorting result of each matched file, and returning the sorting result to the client;
the client is further used for displaying each matched file according to the sorting result;
the process for determining the authority of the issued object in each field comprises the following steps:
determining N target release objects corresponding to files which are recalled by the search requests and clicked corresponding to the strong search intentions within the statistical time period; the strong search intention is a search intention of which the search intention score exceeds a preset intention score;
according to the file issued by the ith target issuing object, the number of times that the strong search intention of the jth field corresponds to the search request and is recalled and clicked and the score of the strong search intention, determining the initial authority of the ith target issuing object in the jth field, wherein i takes a value of [1, N ], j takes a value of [1, M ], and M is the total number of the fields.
13. A sorting apparatus for searching for a file, comprising: at least one memory and at least one processor; the memory stores a program, and the processor calls the program stored in the memory to realize the steps of the method for sorting search files according to any one of claims 1 to 10.
14. A storage medium, characterized in that the storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of ranking search files according to any of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810847629.8A CN109144953B (en) | 2018-07-27 | 2018-07-27 | Search file sorting method, device, equipment, storage medium and search system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810847629.8A CN109144953B (en) | 2018-07-27 | 2018-07-27 | Search file sorting method, device, equipment, storage medium and search system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109144953A CN109144953A (en) | 2019-01-04 |
| CN109144953B true CN109144953B (en) | 2022-02-01 |
Family
ID=64799147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810847629.8A Active CN109144953B (en) | 2018-07-27 | 2018-07-27 | Search file sorting method, device, equipment, storage medium and search system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109144953B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111259272B (en) * | 2020-01-14 | 2023-06-20 | 口口相传(北京)网络技术有限公司 | Search result ordering method and device |
| CN113656574B (en) * | 2021-10-19 | 2022-02-08 | 北京欧应信息技术有限公司 | Method, computing device and storage medium for search result ranking |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520784A (en) * | 2008-02-29 | 2009-09-02 | 富士通株式会社 | Information issuing system and information issuing method |
| CN101782920A (en) * | 2009-12-23 | 2010-07-21 | 中国科学院自动化研究所 | Integrated session environment-oriented information recommendation method |
| CN102737090A (en) * | 2012-03-21 | 2012-10-17 | 袁行远 | Webpage searching result ordering method and device |
| CN102930029A (en) * | 2012-11-07 | 2013-02-13 | 北京网智天元科技有限公司 | Socialized search engine method and system |
| CN103793418A (en) * | 2012-10-31 | 2014-05-14 | 珠海富讯网络科技有限公司 | Search method of real-time vertical search engine for security industry |
| CN107578285A (en) * | 2017-09-04 | 2018-01-12 | 腾讯科技（深圳）有限公司 | Put-on method, device and the electronic equipment of promotion message |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050203924A1 (en) * | 2004-03-13 | 2005-09-15 | Rosenberg Gerald B. | System and methods for analytic research and literate reporting of authoritative document collections |
| WO2006047654A2 (en) * | 2004-10-25 | 2006-05-04 | Yuanhua Tang | Full text query and search systems and methods of use |
-
2018
- 2018-07-27 CN CN201810847629.8A patent/CN109144953B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520784A (en) * | 2008-02-29 | 2009-09-02 | 富士通株式会社 | Information issuing system and information issuing method |
| CN101782920A (en) * | 2009-12-23 | 2010-07-21 | 中国科学院自动化研究所 | Integrated session environment-oriented information recommendation method |
| CN102737090A (en) * | 2012-03-21 | 2012-10-17 | 袁行远 | Webpage searching result ordering method and device |
| CN103793418A (en) * | 2012-10-31 | 2014-05-14 | 珠海富讯网络科技有限公司 | Search method of real-time vertical search engine for security industry |
| CN102930029A (en) * | 2012-11-07 | 2013-02-13 | 北京网智天元科技有限公司 | Socialized search engine method and system |
| CN107578285A (en) * | 2017-09-04 | 2018-01-12 | 腾讯科技（深圳）有限公司 | Put-on method, device and the electronic equipment of promotion message |
Non-Patent Citations (2)
| Title |
|---|
| 基于用户分析的个性化微博推荐技术研究;王梦阳;《中国优秀硕士学位论文全文数据库 信息科技辑》;20160415;第I138-1534页 * |
| 微博搜索的关键技术研究;段亚娟;《中国博士学位论文全文数据库 信息科技辑》;20140915;第I139-5页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109144953A (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424043B (en) | Product recommendation method and device and electronic equipment | |
| WO2020048084A1 (en) | Resource recommendation method and apparatus, computer device, and computer-readable storage medium | |
| US8290927B2 (en) | Method and apparatus for rating user generated content in search results | |
| US9208202B1 (en) | Systems and methods for determining interest in an item or category of items | |
| CN106708817B (en) | Information searching method and device | |
| CN106651544B (en) | Conversational recommendation system with minimal user interaction | |
| US9384278B2 (en) | Methods and systems for assessing excessive accessory listings in search results | |
| CN110175895B (en) | Article recommendation method and device | |
| CN108648058B (en) | Product sorting method and device, electronic equipment and storage medium | |
| US20140358914A1 (en) | Search method, search device and storage medium | |
| CN107808314B (en) | User recommendation method and device | |
| US20210382609A1 (en) | Method and device for displaying multimedia resource | |
| JP5831204B2 (en) | Information providing system, information providing method, and program | |
| CN112084405A (en) | Searching method, searching device and computer storage medium | |
| US9600586B2 (en) | System and method for metadata transfer among search entities | |
| US20170287041A1 (en) | Information processing apparatus, information processing method, and information processing program | |
| CN111259173A (en) | Search information recommendation method and device | |
| CN114820123A (en) | Group purchase commodity recommendation method, device, equipment and storage medium | |
| CN110781307A (en) | Target item keyword and title generation method, search method and related equipment | |
| US8700648B2 (en) | Context based networking | |
| CN108536786A (en) | A kind of information recommendation method, device, server and storage medium | |
| CN109144953B (en) | Search file sorting method, device, equipment, storage medium and search system | |
| CN107943910A (en) | A kind of Individual book based on combinational algorithm recommends method | |
| CN111310046B (en) | Object recommendation method and device | |
| CN113077317A (en) | Item recommendation method, device and equipment based on user data and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |