WO2020052067A1 - Information search method and device - Google Patents

Abstract

An information search method and device. The method comprises: using search information to search in a target information library, so as to acquire a search result set (201); selecting a search result subset from the search result set, and determining, as supplementary search information, a search result which is a target search result in the search result subset (202); executing the following search steps: determining, as a supplementary search result set, the union of a search result for the supplementary search information in the target information library with the search result set (2031); and determining whether the supplementary search result set meets a preset convergence condition (2032) (203); and in response to determining that the supplementary search result set does not meet the preset convergence condition, selecting a supplementary search result subset from the supplementary search result set, determining, as supplementary search information, a supplementary search result belonging to a target search result in the supplementary search result subset, determining the supplementary search result set to be the search result set, and continuing to execute the search steps (204). The method increases the number of search results.

Description

Method and device for searching information

This patent application claims the priority of a Chinese patent application filed on September 12, 2018, with application number 201811060981.3, the applicant being Beijing BYTE Network Technology Co., Ltd., and the invention name "Methods and Devices for Searching Information" The entire application is incorporated herein by reference.

Technical field

Embodiments of the present application relate to the field of computer technology, and in particular, to a method and an apparatus for searching information.

Background technique

In the search process, the search results corresponding to the input search information are generally directly selected according to the search algorithm. In this way, the information obtained after adjusting the search information multiple times may not be directly searchable, so the search results obtained may not be comprehensive.

Summary of the Invention

The embodiments of the present application provide a method and device for searching information.

In a first aspect, an embodiment of the present application provides a method for searching information. The method includes: using the obtained search information to search in a target information database to obtain a search result set; selecting a search result subset from the search result set, And determining the search result that belongs to the target search result in the search result subset as supplementary search information; performing the following search step: determining the union of the search result and the search result set searched by the supplementary search information in the target information database as the supplementary search result set ; Determine whether the supplementary search result set satisfies a preset convergence condition; in response to determining that the supplementary search result set does not meet the preset convergence condition, selecting a supplementary search result subset from the supplementary search result set, and belonging the supplementary search result subset to the target The supplementary search result of the search result is determined as the supplementary search information, the supplementary search result set is determined as the search result set, and the foregoing search step is continued.

In some embodiments, the above search step further includes: in response to determining that the supplementary search result set satisfies the above-mentioned convergence condition, determining a target search result set from the supplementary search result set.

In some embodiments, determining the search results in the search result subset that belong to the target search result as supplementary search information includes: obtaining annotation information of the search results in the search result subset, wherein the annotation information is used to indicate whether the search result is a target Search results; according to the labeled information, the search results in the search result subset that belong to the target search results are determined.

In some embodiments, after using the search information to search in the target information database to obtain the search result set, the method further includes: determining a correlation degree between the search results in the search result set, where the correlation degree is used to represent the relationship between the search result and the search information. Relevance.

In some embodiments, for the search results in the search result set, the relevance of the search results is determined by the following steps: determining the similarity between the search results and the search information; and determining the relevance of the search results according to the similarity, wherein, Similarity is directly proportional to relevance.

In some embodiments, after determining the union of the search results and the search result set of the supplementary search information in the target information database as the supplementary search result set, the method further includes: responding to the supplementary search results in the supplementary search result set, in response to the The supplementary search result exists in the search result set, and the correlation degree of the supplementary search result is updated according to a preset correlation degree improvement algorithm, and the updated correlation degree is used as the correlation degree of the supplementary search result; Exist in the search result set and determine the relevance of the supplementary search result.

In some embodiments, the determining the correlation degree of the supplementary search result includes: determining a similarity between the supplementary search result and the search information; and determining a correlation degree of the supplementary search result according to the similarity, where the similarity is related to the correlation degree Directly proportional.

In some embodiments, the above determining the target search result set from the supplementary search result set includes: selecting the target number of supplementary search results from the supplementary search result set as the target search results in the order of the degree of relevance from the supplementary search result set to obtain the target search. Result set.

In some embodiments, the method further includes: dividing the search result set into corresponding target number search result subsets according to a preset number of target relevance intervals; and selecting the search result subset from the search result set, including : Select search results from the target number of search result subsets to obtain search result subsets.

In some embodiments, the method further comprises: determining an accuracy of the search result subset of the target number of search result subsets, wherein the accuracy is used to represent a proportion of the target search results in the search result subset.

In some embodiments, after determining the union of the search result and the search result set of the supplementary search information in the target information base as the supplementary search result set, the method further includes: dividing the supplementary search result set into Corresponding target number of supplementary search result subsets; and selecting supplementary search result subsets from the supplementary search result set, including: separately selecting supplementary search results from the target number of supplementary search result subsets to obtain a supplementary search result subset.

In some embodiments, the method further comprises: determining an accuracy of the supplementary search result subset of the target number of supplementary search result sets, where the accuracy is used to represent a proportion of target search results in the supplementary search result subset.

In some embodiments, the above-mentioned selecting supplementary search results from the target number of supplementary search result subsets respectively includes: for the supplementary search subset of the target number of supplementary search result subsets, determining the relevance degree of the supplementary search result subset. The absolute value of the difference between the accuracy of the search result set corresponding to the interval and the preset accuracy threshold; based on the absolute value and the number of supplementary search results contained in the supplementary search result subset, determining the sub-search results from the supplementary search result set The number of centrally selected supplementary search results. The absolute value is inversely proportional to the number of supplementary search results selected. The number of supplementary search results included is directly proportional to the number of supplementary search results selected.

In some embodiments, the number of search results in the search result set and the number of supplementary search results in the supplementary search result subset are the same.

In some embodiments, the convergence condition includes: a difference between the accuracy of the search result subset and the supplementary search result subset corresponding to the target relevance degree interval is less than a preset difference threshold.

In some embodiments, the convergence condition includes: the number of search results included in the difference between the search result set and the supplementary search result set is less than a preset number difference threshold.

In a second aspect, an embodiment of the present application provides an apparatus for searching for information. The apparatus includes: a searching unit configured to search in a target information database using the obtained search information to obtain a search result set; and a selecting unit, which is It is configured to select a search result subset from the search result set, and determine the search results in the search result subset that belong to the target search result as supplementary search information; the supplementary search unit is configured to perform the following search steps: placing the supplementary search information in the target information The union of the search results and the search result set searched in the library is determined as the supplementary search result set; determining whether the supplementary search result set satisfies a preset convergence condition; and a determining unit configured to respond to determining that the supplementary search result set does not satisfy the preset Condition of convergence, select a supplementary search result subset from the supplementary search result set, and determine the supplementary search results belonging to the target search result in the supplementary search result subset as supplementary search information, determine the supplementary search result set as the search result set, and continue execution The search steps above.

In some embodiments, the above-mentioned supplementary search unit is further configured to determine a target search result set from the supplementary search result set in response to determining that the supplementary search result set satisfies the above-mentioned convergence condition.

In some embodiments, the above-mentioned selecting unit is further configured to: obtain annotation information of the search results in the search result subset, wherein the annotation information is used to indicate whether the search result is a target search result; and determine the search result subset according to the annotation information Search results that belong to the target search result.

In some embodiments, the apparatus further includes a relevance degree determining unit configured to determine a relevance degree of the search results in the search result set, wherein the relevance degree is used to indicate a relevance degree of the search results and the search information.

In some embodiments, the above-mentioned relevance determination unit is further configured to: determine a similarity between the search result and the search information; and determine a relevance of the search result according to the similarity, wherein the similarity is directly proportional to the relevance.

In some embodiments, the above-mentioned correlation degree determining unit is further configured to: for the supplementary search result set in the supplementary search result set, in response to the supplementary search result existing in the search result set, update the supplement according to a preset correlation degree improvement algorithm. The relevance of the search results and the updated relevance are taken as the relevance of the supplementary search results; in response to the supplementary search results not being present in the search result set, the relevance of the supplementary search results is determined.

In some embodiments, the above-mentioned relevance determination unit is further configured to: determine a similarity between the supplementary search result and the search information; and determine a relevance between the supplementary search results according to the similarity, wherein the similarity is directly proportional to the relevance .

In some embodiments, the above-mentioned supplementary search unit is further configured to select the target number of supplementary search results from the supplementary search result set as the target search results in order of the relevance degree from large to small to obtain the target search result set.

In some embodiments, the apparatus further includes a dividing unit configured to divide the search result set into a corresponding number of target search result subsets according to a preset number of target relevance intervals; and the above-mentioned selecting unit, further It is configured to select the search results from the target number of search result subsets respectively to obtain the search result subsets.

In some embodiments, the apparatus further includes: an accuracy determination unit configured to determine the accuracy of the search result subset in the search result subset of the target number, where the accuracy is used to represent the target search result in the search result subset Percentage.

In some embodiments, the above-mentioned dividing unit is further configured to divide the supplementary search result set into corresponding number of supplementary search result subsets according to the above-mentioned relevance degree interval; and the above-mentioned determination unit is further configured to: from the number of targets The supplementary search result subsets are selected from each of the supplementary search result subsets to obtain the supplementary search result subsets.

In some embodiments, the accuracy determination unit is further configured to determine the accuracy of the supplementary search result subset in the target number of supplementary search result sets, where the accuracy is used to represent the target search result location in the supplementary search result subset. Percentage.

In some embodiments, the above-mentioned determining unit is further configured to: for the supplementary search subset of the target number of supplementary search result subsets, determine the accuracy and An absolute value of a difference between a preset accuracy threshold; determining the number of supplementary search results selected from the supplementary search result subset according to the absolute value and the number of supplementary search results included in the supplementary search result subset, wherein The absolute value is inversely proportional to the number of selected supplementary search results, and the number of supplementary search results included is directly proportional to the number of supplementary search results selected.

In some embodiments, the number of search results in the search result set and the number of supplementary search results in the supplementary search result subset are the same.

In some embodiments, the convergence condition includes: a difference between the accuracy of the search result subset and the supplementary search result subset corresponding to the target relevance degree interval is less than a preset difference threshold.

In some embodiments, the convergence condition includes: the number of search results included in the difference between the search result set and the supplementary search result set is less than a preset number difference threshold.

According to a third aspect, an embodiment of the present application provides an electronic device. The electronic device includes: one or more processors; a storage device configured to store one or more programs; The processor executes such that one or more processors implement the method as described in any implementation of the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer-readable medium having stored thereon a computer program that, when executed by a processor, implements the method as described in any implementation manner of the first aspect.

The method and device for searching information provided in the embodiments of the present application can obtain a search result by directly sampling a part of the search result and selecting a search result belonging to the target search result to perform a search again. After that, according to the search result of the search again Determine whether to continue sampling and searching, so as to increase the number of search results and improve the coverage of search results.

BRIEF DESCRIPTION OF THE DRAWINGS

Other features, objects, and advantages of the present application will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings:

FIG. 1 is an exemplary system architecture diagram to which an embodiment of the present application can be applied; FIG.

2 is a flowchart of an embodiment of a method for searching information according to the present application;

3 is a flowchart of still another embodiment of a method for searching information according to the present application;

4 is a flowchart of another embodiment of a method for searching information according to the present application;

5 is a schematic diagram of an application scenario of a method for searching information according to an embodiment of the present application;

6 is a schematic structural diagram of an embodiment of an apparatus for searching information according to the present application;

FIG. 7 is a schematic structural diagram of a computer system suitable for implementing an electronic device according to an embodiment of the present application.

detailed description

The following describes the present application in detail with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention, rather than limiting the invention. It should also be noted that, for the convenience of description, only the parts related to the related invention are shown in the drawings.

It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The application will be described in detail below with reference to the drawings and embodiments.

FIG. 1 illustrates an exemplary architecture 100 of an embodiment of a method for searching information or a device for searching information to which the present application can be applied.

As shown in FIG. 1, the system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, and so on.

The terminal devices 101, 102, 103 interact with the server 105 through the network 104 to receive or send messages and the like. Various client applications can be installed on the terminal devices 101, 102, and 103, such as a web browser application, a shopping application, a search application, and an instant communication tool.

The terminal devices 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, and 103 are hardware, they can be various electronic devices that support data storage and data exchange, including but not limited to smartphones, tablets, e-book readers, laptop computers, and desktop computers. Wait. When the terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (such as multiple software or software modules used to provide distributed services), or it can be implemented as a single software or software module. It is not specifically limited here.

The server 105 may be a server that provides various services, for example, a search server that performs a search in the target information base according to the search information sent by the terminal devices 101, 102, and 103, and returns search results to the terminal devices 101, 102, and 103.

It should be noted that the above search information and target information base may also be directly stored locally on the server 105, and the server 105 may directly extract the search information and search in the target information base stored locally. At this time, there may be no terminal Devices 101, 102, 103 and network 104.

It should be noted that the method for searching information provided by the embodiment of the present application is generally executed by the server 105, and accordingly, the apparatus for searching information is generally set in the server 105.

It should also be noted that information search applications can also be installed in the terminal devices 101, 102, and 103. At this time, the method for processing images can also be executed by the terminal devices 101, 102, and 103, and accordingly, used for processing The image device may be provided in the terminal devices 101, 102, and 103. At this time, the exemplary system architecture 100 may be absent from the server 105 and the network 104.

It should be noted that the server may be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers or as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, to provide distributed services), or it can be implemented as a single software or software module. It is not specifically limited here.

It should be understood that the numbers of terminal devices, networks, and servers in FIG. 1 are merely exemplary. According to implementation needs, there can be any number of terminal devices, networks, and servers.

With continued reference to FIG. 2, a flowchart 200 of an embodiment of a method for searching information according to the present application is shown. The method for searching information includes the following steps:

Step 201: Use the obtained search information to search in a target information database to obtain a search result set.

In this embodiment, the execution subject of the method for searching information (such as the server 105 in FIG. 1) may first obtain the search information from a local or other storage device, and then perform a search in the target information base to obtain a search result set. The target information base may be an information base specified by a user in advance, or a search base matching the search information determined according to the search information, or an information base set according to actual application requirements. The search algorithm used in the search process may be preset.

Step 202: Select a search result subset from the search result set, and determine search results belonging to the target search result in the search result subset as supplementary search information.

In this embodiment, the search result set may be sampled first, and at least one search result may be selected to obtain a search result subset. The sampling ratio can be set randomly, or preset by the user or technician, or it can be determined based on the size of the search result set (for example, ten percent of the number of search results included in the search result set is selected as Number of search result subsets).

In this embodiment, the target search result may refer to a search result that meets a user's search needs. For example, it is possible to verify whether each search result is a target search result by calculating the similarity between each search result in the search result subset and the search information.

Optionally, the annotation information of the search results in the search result subset may also be obtained first, where the annotation information is used to indicate whether the search result is a target search result. Then, according to the label information, search results in the search result subset that belong to the target search result are determined.

Specifically, each search result in the search result subset can be manually verified, and the verification result is input as labeled information, and then sent to the above-mentioned execution subject.

Step 203: Perform the following search steps:

In step 2031, the union of the search result and the search result set searched by the supplementary search information in the target information database is determined as the supplementary search result set.

In this embodiment, a search may be performed in the target information database by using the supplementary search information to obtain a search result. Because supplementary search information is used, compared with the above search result set, new search results may be included in this search result. Therefore, the union of the current search result and the search result set can be determined as a supplementary search result set.

Step 2032: Determine whether the supplementary search result set satisfies a preset convergence condition.

In this embodiment, the convergence condition may be specifically set by a user or a technician according to an actual search requirement. For example, the convergence condition may be that the number of supplementary search result sets is greater than a preset threshold. For another example, the convergence condition may also be that the number of search results included in the difference between the supplementary search result set and the search result set is less than a preset threshold.

Step 204: In response to determining that the supplementary search result set does not satisfy a preset convergence condition, selecting a supplementary search result subset from the supplementary search result set, and determining the supplementary search result belonging to the target search result in the supplementary search result subset as the supplementary search information. , Determine the supplementary search result set as the search result set, and continue with the above search steps.

In this embodiment, when the supplementary search result set does not satisfy a preset convergence condition, a method similar to step 202 above may be followed, and the supplementary search result set may be sampled first to obtain a supplementary search result subset, and then the supplementary search is performed. The supplementary search results belonging to the target search result in the subset are determined as supplementary search information, the supplementary search result set is determined as the search result set, and the above step 203 is continuously performed to perform an iterative search.

Optionally, in response to determining that the supplementary search result set satisfies the above-mentioned convergence condition, a target search result set is determined from the supplementary search result set. For example, the supplementary search result set can be directly used as the target search result set.

The method for searching information provided by the above embodiments of the present application samples the search result set obtained by searching the search information in the target information database, and uses the search results in the sampled search result subset that belong to the target search result as a supplementary search. The information is searched again in the target information base to supplement the new search results. After that, you can further determine whether the supplementary search result set obtained by adding the new search result to the search result set meets the convergence conditions. If it is not satisfied, you can also iteratively search in the target information database according to the method of sampling and supplementary search information described above, so as to continuously Add new search results until all the search results obtained meet the convergence conditions, thereby improving the coverage of the search results and avoiding the situation where a direct search may miss many search results.

With further reference to FIG. 3, a flowchart 300 of yet another embodiment of a method for searching information is shown. The process 300 of the method for searching information includes the following steps:

Step 301: Use the obtained search information to search in a target information database to obtain a search result set.

For the specific execution process of this step, reference may be made to the related description of step 201 in the embodiment corresponding to FIG. 2, and details are not described herein again.

Step 302: Determine the relevance of the search results in the search result set.

In this embodiment, the degree of association may be used to indicate the degree of association between the search results and the search information. Specifically, a correlation degree between each search result and the search information may be calculated according to a preset correlation degree calculation method, or a correlation degree between each search result and the search information may be manually marked and returned to the above-mentioned execution subject.

Optionally, for the search results in the search result set, the relevance of the search results may be determined by the following steps: first, the existing similarity calculation methods may be used to determine the similarity between the search result and the search information, and then according to the similarity The relationship is proportional to the degree of relevance, and the degree of relevance of the search result is determined based on the similarity. For example, the similarity may be directly determined as the relevance of the search result.

Step 303: Select a search result subset from the search result set, and determine search results belonging to the target search result in the search result subset as supplementary search information.

For the specific execution process of this step, reference may be made to the description of step 202 in the embodiment corresponding to FIG. 2, and details are not described herein again.

Step 304: Perform the following search steps:

Step 3041: Determine the union of the search result and the search result set searched in the target information base by the supplementary search information as the supplementary search result set.

For the specific execution process of step 3041, reference may be made to the description of step 2031 in the embodiment corresponding to FIG. 2, and details are not described herein again.

Step 3042: For the supplementary search results in the supplementary search result set, determine the relevance of the supplementary search results in the following manner:

Step 30421: In response to the supplementary search result existing in the search result set, update the correlation degree of the supplementary search result according to a preset correlation degree improvement algorithm, and use the updated correlation degree as the correlation degree of the supplementary search result.

In this embodiment, for the supplementary search results existing in the search result set, that is, the search results are searched again, the relevance of the supplementary search results can be improved. Specifically, a technician can set a correlation improvement algorithm in advance to improve the correlation, and can also set a correlation improvement value in advance. When a search result is searched again, the correlation of the search result is added to the set correlation. Value, and use the obtained new relevance as the relevance of this search result.

Step 30422, in response to the supplementary search result not existing in the search result set, determining the relevance of the supplementary search result.

In this embodiment, some existing similarity calculation methods may be used to first determine the similarity between the supplementary search result and the search information, and then determine the supplementary search result based on the similarity according to the relationship between the similarity and the correlation. Correlation. For example, the similarity may be directly determined as the relevance of the supplementary search result.

For a supplementary search result that does not exist in the search result set, that is, a newly added search result, the specific calculation process of the degree of association may also refer to the related description of step 302, which is not repeated here.

Step 3043: Determine whether the supplementary search result set satisfies a preset convergence condition.

For the specific execution process of step 3043, refer to the related description of step 2032 in the embodiment corresponding to FIG. 2, and details are not described herein again.

Step 305: In response to determining that the supplementary search result set does not satisfy a preset convergence condition, select a supplementary search result subset from the supplementary search result set, and determine the supplementary search result belonging to the target search result in the supplementary search result subset as the supplementary search information. , Determine the supplementary search result set as the search result set, and continue with the above search steps.

For the specific execution process of this step, reference may be made to the related description of step 204 in the embodiment corresponding to FIG. 2, and details are not described herein again.

In step 306, in response to determining that the supplementary search result set satisfies the above-mentioned convergence condition, the target search result set is selected from the supplementary search result set as the target search result in the order of the correlation degree from large to small to obtain a target search result set.

In this embodiment, the number of targets may be set in advance, or may be determined according to a number determination rule (for example, the number of targets is half of the number of supplementary search results included in the supplementary search result set).

It can be seen from FIG. 3 that, compared with the embodiment corresponding to FIG. 2, the process for searching information in this embodiment can also determine the degree of relevance corresponding to each search result for the search results after each search. During the iterative search process, the relevance of search results that are searched again can be dynamically updated. Therefore, the target search result can be selected according to the relevance of each search result and the supplementary search result, thereby further improving the accuracy of the search result.

With further reference to FIG. 4, a flowchart 400 of still another embodiment of a method for searching information is shown. The process 400 of the method for searching information includes the following steps:

Step 401: Use the obtained search information to search in a target information database to obtain a search result set.

For the specific execution process of this step, reference may be made to the related description of step 201 in the embodiment corresponding to FIG. 2, and details are not described herein again.

Step 402: Determine the relevance of the search results in the search result set.

For the specific execution process of this step, reference may be made to the description of step 302 in the embodiment corresponding to FIG. 3, and details are not described herein again.

Step 403: Divide the search result set into a corresponding number of target search result subsets according to a preset number of target relevance intervals.

In this embodiment, the target number may be preset according to actual application requirements. Among them, each correlation degree interval may be evenly divided or non-uniformly divided. For example, for the case where the correlation degree is between 0 and 1 (including 0 and 1), it can be divided into the following five correlation degree intervals: 0 to 0.2 (including 0, not including 0.2), 0.2 to 0.4 ( Including 0.2, excluding 0.4), 0.4 to 0.6 (including 0.4, excluding 0.6), 0.6 to 0.8 (including 0.6, excluding 0.8) to 0.8-1 (including 0.8 and 1), and it can also be non-uniformly divided into the following: Five correlation intervals: 0 to 0.5 (including 0, not including 1), 0.5 to 0.7 (including 0.5, not including 0.7), 0.7 to 0.8 (including 0.7, not including 0.8), 0.8 to 0.9 (including 0.8, not Contains 0.9), 0.9 to 1 (including 0.9 and 1).

Taking the above evenly divided five correlation degree intervals as an example, correspondingly, a search result with a correlation degree between 0 and 0.2 can be determined as a search result subset, and a search result with a correlation degree between 0.2 and 0.4 can be determined as A search result subset. A search result with a relevance between 0.4 and 0.6 is determined as a search result subset. A search result with a relevance between 0.6 and 0.8 is determined as a search result subset. Search results between 0.8 and 1 are determined as a subset of search results.

Step 404: Select search results from the target number of search result subsets to obtain the search result subsets, and determine the search results belonging to the target search results in the search result subsets as supplementary search information.

In this embodiment, a partial search result may be sampled from each search result subset, and the sampled search results may be combined to obtain a search result subset. The number of search results sampled for each search result subset may be arbitrarily specified, or may be determined according to a sampling number determination rule. For example, the number of samples determination rule may sample the number of search results for each search result subset to be one tenth of the number of search results included in each search result subset. For another example, the number of samples can also be determined according to the degree of correlation. For example, the closer the corresponding degree of correlation is to 0.5, the larger the number of corresponding samples is.

Optionally, the accuracy of the search result subset in the target number of search result subsets may be further determined. Among them, accuracy can be used to represent the proportion of target search results in the search result subset. For example, for a search result subset containing 30 search results, 20 of which belong to the target search result, the accuracy of the search result subset is two thirds.

Step 405: Perform the following search steps:

Step 4051: Determine the union of the search result and the search result set searched in the target information base by the supplementary search information as the supplementary search result set.

For the specific execution process of step 4051, reference may be made to the description of step 2031 in the embodiment corresponding to FIG. 2, and details are not described herein again.

Step 4052: For the supplementary search results in the supplementary search result set, determine the relevance of the supplementary search results in the following manner:

In step 40521, in response to the supplementary search result existing in the search result set, the correlation degree of the supplementary search result is updated according to a preset correlation degree improvement algorithm, and the updated correlation degree is used as the correlation degree of the supplementary search result.

Step 40522, in response to the supplementary search result not existing in the search result set, determining the relevance of the supplementary search result.

For specific implementation processes of the above steps 40521 and 40522, reference may be made to the related description of steps 30421 and 30422 in the embodiment corresponding to FIG. 3, and details are not described herein again.

In step 4053, the supplementary search result set is divided into a corresponding number of target supplementary search result subsets according to the above-mentioned association degree interval.

In this embodiment, after the supplementary search result set is obtained, new search results may be included, and the relevance of the search results that are searched again is improved, so the supplementary search results may be divided again according to the relevance. For the specific division process, refer to the related description in step 403 above.

Step 4054: Determine whether the supplementary search result set satisfies a preset convergence condition.

For the specific execution process of step 4054, refer to the related description of step 2032 in the embodiment corresponding to FIG. 2, and details are not described herein again.

Step 406: In response to determining that the supplementary search result set does not satisfy a preset convergence condition, select supplementary search results from the target number of supplementary search result sets to obtain a supplementary search result subset, and belong to the target search result to the supplementary search result subset. The supplementary search result of is determined as the supplementary search information, the supplementary search result set is determined as the search result set, and the foregoing search step is continued.

In this embodiment, the search results can be supplemented from the sampling part of each supplementary search result subset, and the sampled supplementary search results can be combined to obtain a supplementary search result subset. The number of supplementary search results sampled for each supplementary search result subset may be arbitrarily specified, or may be determined according to a determination rule of the number of samples. For example, the sampling number determination rule may sample the number of supplementary search results for each supplementary search result subset to be one tenth of the number of supplementary search results included in each supplementary search result subset. For another example, the number of samples can also be determined according to the degree of correlation. For example, the closer the corresponding degree of correlation is to 0.5, the larger the number of corresponding samples is.

Optionally, for the supplementary search subset of the target number of supplementary search result subsets, the accuracy between the search result set corresponding to the correlation interval in which the supplementary search result subset is located and the preset accuracy threshold may be determined. The absolute value of the difference between the two; then, based on the determined absolute value and the number of supplementary search results contained in the supplementary search result subset, the number of supplementary search results selected from the supplementary search result subset is determined, where the determined absolute The value is inversely proportional to the number of selected supplementary search results, and the number of included supplementary search results is directly proportional to the number of selected supplementary search results.

The preset accuracy threshold may be preset by a technician. In practice, the minimum and maximum values of accuracy can be estimated in advance, and the accuracy threshold can be set according to the minimum and maximum values of accuracy that are estimated in advance. For example, if the accuracy is between 0 and 1, then the accuracy threshold can be set to 0.5. At this time, the closer the corresponding accuracy is to the supplementary search result subset of 0.5, the more the number of corresponding samples is; on the contrary, the closer the corresponding accuracy is to the supplementary search result subset of 0 and 1, the more the corresponding number of samples is. less.

For example, the number of supplementary search results included in the two supplementary search result subsets is 200 and 400 respectively, then the number of samples of the supplementary search result subset including 400 supplementary search results may be greater than that of the 200 supplementary search results. The number of samples of the supplementary search result subset.

Optionally, the number of search results in the search result subset and the number of search results in the supplementary search result subset may be the same. That is, after each search, the total number of samples can be fixed.

Optionally, the accuracy of the supplementary search result subset in the target number of supplementary search result sets may be further determined, where the accuracy may be used to represent the proportion of target search results in the supplementary search result subset.

Optionally, the convergence condition may be that the difference between the accuracy of the search result subset and the supplementary search result subset corresponding to the target relevance degree interval is less than a preset difference threshold.

Optionally, the convergence condition may also be that the number of search results included in the difference between the search result set and the supplementary search result set is less than a preset number difference threshold.

In step 407, in response to determining that the supplementary search result set satisfies the above-mentioned convergence conditions, the target search result set is selected from the supplementary search result set as the target search result in the order of the correlation degree from large to small to obtain a target search result set.

For the specific execution process of this step, reference may be made to the description of step 306 in the embodiment corresponding to FIG. 3, and details are not described herein again.

With continued reference to FIG. 5, FIG. 5 is a schematic diagram of an application scenario of the method for searching information according to this embodiment. In the application scenario of FIG. 5, a search result set 503 may be obtained by first searching the image library 502 using the image 501, where the search result set 503 includes 100 images. After that, the similarity between each image in the search result set 503 and the image 501 can be calculated separately, and the similarity can be used as the correlation between each image. After that, the search result set may be divided into three subsets according to the relevance degree (as shown by reference numeral 504 in the figure). The correlation degree of the images in the first subset is between 0 and 0.5, which contains a total of 30 images. The correlation degree of the images in the second subset is between 0.5 and 0.8, which contains a total of 50 images. The relevance of the images is between 0.8 and 1 and contains a total of 20 images.

Then, sampling can be performed in three subsets, and the target images included in the sampling are combined to obtain a supplementary search image 506. Specifically, 10 images can be sampled for the first subset, 10 images can be sampled for the second subset, and 3 images can be sampled for the third subset. Specifically, as shown by reference numeral 505 in the figure, the image sampled by the first subset contains 5 target images, the image sampled by the second subset contains 8 target images, and the image sampled by the third subset contains Since three target images are obtained, 16 target images can be obtained as the supplementary search image 506. Further, it can also be obtained that the accuracy of the 10 images sampled by the first subset is one-half, the accuracy of the sampling by the second subset is four-fifths, and the accuracy by the third subset is one.

Then, the supplementary search image 506 may be searched in the image library 502 to obtain the current search result set 507, where the current search result set 507 includes 120 images. Further, the union of the search result set 503 obtained from the previous search and the supplementary search result set of the current search may be used as the supplementary search result set 508, where the supplementary search result set includes 150 images.

It can be seen that during this search, there were 30 newly searched images and another 90 images were searched again. For 90 images that are searched again, the degree of correlation can be improved. For 30 newly searched images, the similarity between these images and the image 501 can be used as the correlation of these images. Thus, the 150 images included in the supplementary search result set are regrouped according to the relevance degree, and new three subsets are obtained (as shown in 509 in the figure).

The first subset with newly obtained correlation between 0 and 0.5 contains 45 images, the second subset with newly obtained correlation between 0.5 and 0.8 contains 80 images, and the newly obtained correlation with 0.8 The third subset between 1 and 25 contains 25 images.

After that, according to the accuracy of the subset sampling corresponding to the three correlation degree intervals calculated last time, the three newly obtained subsets are sampled again. The accuracy of the last three subsets was one-half, four-fifths, and one, so in this sampling process, the sampling in each subset can be determined according to the preset accuracy threshold (such as 0.5). number.

For the newly obtained first subset, since the absolute value of the difference between the accuracy and the accuracy threshold of the subset is 0, more images can be sampled in the subset. For the newly obtained second subset and the newly obtained third subset, since the absolute value of the difference between the accuracy and the accuracy threshold of the two subsets is large, so fewer images can be sampled in these two subsets . Specifically, since the absolute value of the difference between the accuracy and the accuracy threshold of the newly obtained second subset is smaller than the absolute value of the difference between the accuracy and the accuracy threshold of the newly obtained third subset, The number of samples obtained in the second subset may be greater than the number of samples in the newly obtained third subset.

Specifically, 12 images can be sampled in the newly obtained first subset, 10 images can be sampled in the newly obtained second subset, and 1 image can be sampled in the newly obtained third subset. Specifically, as shown by reference numeral 510 in the figure, each of the images obtained in the newly obtained first subset and the second subset includes 8 target images, and the images obtained in the newly obtained third subset include There are 1 target images. It can further be calculated that the accuracy of the newly obtained first subset is two thirds, the accuracy of the newly obtained second subset is four fifths, and the accuracy of the newly obtained third subset is one. .

After that, the above-mentioned search process is performed again using the supplementary search result set 511 composed of a total of 17 newly sampled images as a supplementary search image, and using the supplementary search result set 508 as a search result set.

Specifically, a search is performed in the image library 502 by using the supplementary search result set 511, and a union of the obtained search result and the supplementary search result set 508 is used as a new supplementary search result set 512. The new supplementary search result set 512 contains 160 images. Similarly, the degree of relevance of the image searched again can be improved. For the newly searched image, the similarity between the newly searched image and the image 501 can be used as the relevance of the newly searched image. After that, the new supplementary search result set 512 is re-divided into three subsets according to the re-determined relevance degree. Specifically, as shown by reference numeral 513 in the figure, the newly divided first subset contains 45 images, the newly divided second subset contains 87 images, and the newly divided third subset contains 28 images.

Because the accuracy of the three correlation intervals calculated last time is two-thirds, four-fifths, and one. The images can be sampled from the three newly divided subsets and the accuracy of this sampling can be calculated. Among them, more images can be sampled for the first subset. The accuracy of the newly divided three subsets is calculated to be one-half, nine-tenths, and one, respectively. It can be seen that the accuracy of the newly divided second subset and the newly divided third subset is small, and the new supplementary search result set 512 is compared with the previous supplementary search result set 508. There are also fewer newly searched images, so a total of 115 images in the newly divided second subset and the newly divided third subset can be directly used as the target image set 514.

As can be seen from FIG. 4, compared with the embodiments corresponding to FIGS. 2 and 3, the flow of the method for searching information in this embodiment highlights that after each search is completed and the degree of relevance of each search result is determined The search results can also be divided into corresponding multiple subsets according to a preset correlation degree interval, and supplementary search is sampled in each subset, so that the supplementary search process can cover each correlation degree interval, and further improve the search result. Coverage. On this basis, after each subset is sampled separately, the accuracy corresponding to each subset can be further determined according to the sampling result, and the number of samples for the next interval of the correlation degree can be determined according to the accuracy, thereby achieving targeted Perform supplementary searches to further improve the accuracy and search efficiency of search results.

With further reference to FIG. 6, as an implementation of the methods shown in the foregoing figures, this application provides an embodiment of a device for searching for information. The device embodiment corresponds to the method embodiment shown in FIG. 2, and the device is specific Can be applied to various electronic devices.

As shown in FIG. 6, the apparatus 600 for searching information provided in this embodiment includes a search unit 601, a selection unit 602, a supplemental search unit 603, and a determination unit 604. The search unit 601 is configured to search the target information base using the obtained search information to obtain a search result set; the selection unit 602 is configured to select a search result subset from the search result set, and the search result subset belongs to The search result of the target search result is determined as the supplementary search information; the supplemental search unit 603 is configured to perform the following search step: determining the union of the search result and the search result set searched by the supplementary search information in the target information base as the supplementary search result Determining whether the set of supplementary search results meets a preset convergence condition; a determining unit 604 configured to select a supplementary search result subset from the set of supplemental search results in response to determining that the set of supplemental search results does not meet the preset convergence conditions, and The supplementary search results belonging to the target search result subset in the supplementary search result subset are determined as supplementary search information, the supplementary search result set is determined as the search result set, and the foregoing search step is continued.

In this embodiment, in the apparatus 600 for searching information: the specific processing of the search unit 601, the selection unit 602, the supplemental search unit 603, and the determination unit 604 and the technical effects brought by them can be referred to the corresponding embodiment in FIG. 2 respectively. Relevant descriptions of step 201, step 202, step 203, and step 204 in the description are not repeated here.

In some optional implementations of this embodiment, the above-mentioned supplementary search unit 603 is further configured to: in response to determining that the supplementary search result set satisfies the above-mentioned convergence condition, determine a target search result set from the supplementary search result set.

In some optional implementation manners of this embodiment, the above-mentioned selecting unit 602 is further configured to: obtain annotation information of the search results in the search result subset, wherein the annotation information is used to indicate whether the search result is a target search result; according to Annotate the information to determine the search results that belong to the target search result in the search result subset.

In some optional implementation manners of this embodiment, the apparatus 600 for searching information further includes: a relevance determination unit (not shown in the figure) configured to determine a relevance of search results in a search result set, where , The degree of relevance is used to indicate the degree of relevance of the search results to the search information.

In some optional implementation manners of this embodiment, the above-mentioned relevance determination unit is further configured to: determine a similarity between the search result and the search information; and determine a relevance between the search results according to the similarity, where the similarity Proportionally related.

In some optional implementations of this embodiment, the above-mentioned correlation degree determining unit is further configured to: for a supplementary search result in a supplementary search result set, in response to the supplementary search result existing in the search result set, according to a preset association Degree improvement algorithm, updating the relevance degree of the supplementary search result, and using the updated relevance degree as the relevance degree of the supplementary search result; determining the relevance degree of the supplementary search result in response to the supplementary search result not existing in the search result set .

In some optional implementation manners of this embodiment, the above-mentioned correlation degree determining unit is further configured to: determine a similarity between the supplementary search result and the search information; and determine a correlation degree of the supplementary search result according to the similarity, where: Similarity is directly proportional to relevance.

In some optional implementation manners of this embodiment, the above-mentioned supplementary search unit 603 is further configured to: select the number of supplementary search results from the supplementary search result set as the target search result in the order of relevance from large to small, Get the target search result set.

In some optional implementation manners of this embodiment, the apparatus 600 for searching information further includes a dividing unit (not shown in the figure) configured to: search results according to a preset number of target relevance intervals. The set is divided into a corresponding number of target search result subsets; and the above-mentioned selection unit is further configured to: respectively select search results from the target number of search result subsets to obtain a search result subset.

In some optional implementation manners of this embodiment, the apparatus 600 for searching information further includes: an accuracy determination unit (not shown in the figure) configured to determine search result sub-sets in the target number search result subset. Set accuracy, where accuracy is used to represent the proportion of target search results in a subset of search results.

In some optional implementations of this embodiment, the above-mentioned dividing unit is further configured to divide the supplementary search result set into corresponding target number supplementary search result subsets according to the above-mentioned correlation degree interval; and the above-mentioned determining unit further It is configured to select supplementary search results from the target number of supplementary search result subsets respectively to obtain a supplementary search result subset.

In some optional implementation manners of this embodiment, the accuracy determination unit is further configured to determine the accuracy of the supplementary search result subset of the target number of supplementary search result sets, where the accuracy is used to represent the supplementary search results The percentage of target search results in the subset.

In some optional implementation manners of this embodiment, the foregoing determining unit is further configured to: for a supplementary search subset of the target number of supplementary search result subsets, determine a correlation degree interval corresponding to the supplementary search result subset. The absolute value of the difference between the accuracy of the search result set and a preset accuracy threshold; based on the absolute value and the number of supplementary search results contained in the supplementary search result subset, determining the selected from the supplementary search result subset The number of supplementary search results, where the absolute value is inversely proportional to the number of selected supplementary search results, and the number of supplementary search results included is directly proportional to the number of selected supplementary search results.

In some optional implementation manners of this embodiment, the number of search results in the search result set and the number of supplementary search results in the supplementary search result subset are the same.

In some optional implementation manners of this embodiment, the convergence conditions include: a difference between the accuracy of the search result subset and the supplementary search result subset corresponding to the target relevance degree interval is less than a preset difference threshold.

In some optional implementation manners of this embodiment, the convergence condition includes: the number of search results included in the difference between the search result set and the supplementary search result set is less than a preset number difference threshold.

In the device provided by the foregoing embodiment of the present application, after the search result is directly obtained by the search unit, the partial search result may be sampled by the selection unit and the search result belonging to the target search result may be selected and searched again by the supplementary search unit. The search result of searching again determines whether to continue sampling for searching, thereby achieving an increase in the number of search results and improving the coverage of the search results.

Reference is now made to FIG. 7, which illustrates a schematic structural diagram of a computer system 700 suitable for implementing an electronic device according to an embodiment of the present application. The electronic device shown in FIG. 7 is only an example, and should not impose any limitation on the functions and scope of use of the embodiments of the present application.

As shown in FIG. 7, the computer system 700 includes a central processing unit (CPU) 701, which can be loaded into a random access memory (RAM) 703 from a program stored in a read-only memory (ROM) 702 or from a storage section 708. Instead, perform various appropriate actions and processes. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, ROM 702, and RAM 703 are connected to each other through a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

The following components are connected to the I / O interface 705: an input portion 706 including a keyboard, a mouse, and the like; an output portion 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage portion 708 including a hard disk and the like And a communication section 709 including a network interface card such as a LAN card, a modem, and the like. The communication section 709 performs communication processing via a network such as the Internet. The driver 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read out therefrom is installed into the storage section 708 as needed.

In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product including a computer program carried on a computer-readable medium, the computer program containing program code for performing a method shown in a flowchart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 709, and / or installed from a removable medium 711. When the computer program is executed by a central processing unit (CPU) 701, the above-mentioned functions defined in the method of the present application are executed.

It should be noted that the computer-readable medium of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the foregoing. The computer-readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable Programming read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing. In this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in combination with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal that is included in baseband or propagated as part of a carrier wave, and which carries computer-readable program code. Such a propagated data signal may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, and the computer-readable medium may send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device . Program code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, which contains one or more functions to implement a specified logical function Executable instructions. It should also be noted that in some alternative implementations, the functions noted in the blocks may also occur in a different order than those marked in the drawings. For example, two successively represented boxes may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified function or operation , Or it can be implemented with a combination of dedicated hardware and computer instructions.

The units described in the embodiments of the present application may be implemented by software or hardware. The described unit may also be provided in the processor, for example, it may be described as: a processor including a search unit, a selection unit, a supplementary search unit, and a determination unit. Among them, the names of these units do not constitute a limitation on the unit itself in some cases. For example, the search unit can also be described as "a unit that uses the obtained search information to search in the target information base to obtain a search result set." .

As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the foregoing embodiments; or may exist alone without being assembled into the electronic device in. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: searches the target information base using the obtained search information to obtain a search result set; Select a search result subset in the search result set, and determine the search results that belong to the target search result in the search result subset as supplementary search information; perform the following search steps: search results and search result sets that search for the supplementary search information in the target information database The union set is determined as the supplementary search result set; determining whether the supplementary search result set satisfies a preset convergence condition; and in response to determining that the supplementary search result set does not satisfy the preset convergence condition, selecting a supplementary search result subset from the supplementary search result set, And determining the supplementary search result belonging to the target search result in the supplementary search result subset as supplementary search information, determining the supplementary search result set as the search result set, and continuing to execute the search step described above.

The above description is only a preferred embodiment of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to the technical solution of the specific combination of the above technical features, but it should also cover the above technical features or Other technical solutions formed by arbitrarily combining their equivalent features. For example, a technical solution formed by replacing the above features with technical features disclosed in the present application (but not limited to) with similar functions.

Claims (34)

1. A method for searching information includes:

   Use the obtained search information to search in the target information database to obtain a search result set;

   Selecting a search result subset from the search result set, and determining search results belonging to the target search result in the search result subset as supplementary search information;

   Perform the following search steps: determine the union of the search result and the search result set searched in the target information base by the supplementary search information as the supplementary search result set; determine whether the supplementary search result set satisfies a preset convergence condition;

   In response to determining that the supplementary search result set does not satisfy a preset convergence condition, selecting a supplementary search result subset from the supplementary search result set, and determining the supplementary search result belonging to the target search result in the supplementary search result subset as supplementary search information, and supplementing the The search result set is determined as the search result set, and the search step is continued.
2. The method according to claim 1, wherein the searching step further comprises:

   In response to determining that the supplementary search result set satisfies the convergence condition, a target search result set is determined from the supplementary search result set.
3. The method according to claim 1, wherein determining the search results belonging to the target search result in the search result subset as supplementary search information comprises:

   Obtain annotation information of search results in a search result subset, where the annotation information is used to indicate whether the search result is a target search result;

   According to the labeled information, the search results in the search result subset that belong to the target search result are determined.
4. The method according to claim 2, wherein after said using the search information to search in a target information database to obtain a search result set, further comprising:

   Determine the relevance of the search results in the search result set, where the relevance is used to indicate the relevance of the search results to the search information.
5. The method according to claim 4, wherein, for the search results in the search result set, the relevance of the search results is determined by the following steps:

   Determine the similarity between the search results and the search information;

   According to the similarity, the relevance of the search result is determined, wherein the similarity is directly proportional to the relevance.
6. The method according to claim 4, wherein after determining the union of the search result and the search result set of the supplementary search information in the target information base as the supplementary search result set, further comprising:

   For the supplementary search results in the supplementary search result set, in response to the supplementary search results existing in the search result set, according to a preset correlation degree improvement algorithm, update the correlation degree of the supplementary search results, and use the updated correlation degree as the supplement The relevance of the search results; in response to the supplementary search results not being present in the search result set, the relevance of the supplementary search results is determined.
7. The method according to claim 6, wherein the determining the relevance of the supplementary search result comprises:

   Determine the similarity between the supplementary search results and the search information;

   The relevance degree of the supplementary search result is determined according to the similarity degree, wherein the similarity degree is directly proportional to the relevance degree.
8. The method according to claim 6, wherein said determining a target search result set from a supplementary search result set comprises:

   According to the order of relevance, from the supplementary search result set, the target number of supplementary search results are selected as the target search result, and the target search result set is obtained.
9. The method according to claim 4, wherein the method further comprises:

   Dividing the search result set into a corresponding number of target search result subsets according to a preset number of target relevance intervals; and

   The selecting a search result subset from the search result set includes:

   The search results are selected from the target number of search result subsets to obtain the search result subsets.
10. The method according to claim 9, wherein the method further comprises:

    Determine the accuracy of the search result subset in the number of targets search result subset, where the accuracy is used to represent the proportion of the target search results in the search result subset.
11. The method according to claim 10, wherein after determining the union of the search result and the search result set of the supplementary search information in the target information base as the supplementary search result set, further comprising:

    Dividing the supplementary search result set into corresponding target number supplementary search result subsets according to the correlation degree interval; and

    The selecting a supplementary search result subset from the supplementary search result set includes:

    The supplementary search results are selected from the target number of supplementary search result subsets to obtain the supplementary search result subsets.
12. The method according to claim 11, wherein the method further comprises:

    Determine the accuracy of the supplementary search result subset of the target number of supplementary search result sets, where the accuracy is used to represent the proportion of target search results in the supplementary search result subset.
13. The method according to claim 12, wherein the selecting the supplementary search results from the target number of supplementary search result subsets respectively comprises:

    For the supplementary search subset of the target number of supplementary search result subsets, determine the absolute value of the difference between the accuracy of the search result set corresponding to the correlation interval in which the supplementary search result subset is located and the preset accuracy threshold. Determining the number of supplementary search results selected from the supplementary search result subset according to the absolute value and the number of supplementary search results included in the supplementary search result subset, wherein the absolute value and the number of supplementary search results selected The number is inversely proportional, and the number of supplementary search results included is proportional to the number of selected supplementary search results.
14. The method according to any one of claims 1-13, wherein the number of search results in the search result set and the number of supplementary search results in the supplementary search result subset are the same.
15. The method according to claim 13, wherein the convergence conditions include:

    The difference between the accuracy of the search result subset and the supplementary search result subset corresponding to the target relevance degree interval is less than a preset difference threshold.
16. The method according to claim 1, wherein the convergence conditions include:

    The number of search results included in the difference between the search result set and the supplementary search result set is less than a preset number difference threshold.
17. A device for searching information includes:

    A search unit configured to search the target information base using the obtained search information to obtain a search result set;

    A selecting unit configured to select a search result subset from the search result set, and determine search results belonging to the target search result in the search result subset as supplementary search information;

    The supplementary search unit is configured to perform the following search steps: determining the union of the search result and the search result set searched by the supplementary search information in the target information base as the supplementary search result set; determining whether the supplementary search result set satisfies a preset convergence condition;

    A determining unit configured to, in response to determining that the supplementary search result set does not satisfy a preset convergence condition, select a supplementary search result subset from the supplementary search result set, and determine the supplementary search result belonging to the target search result in the supplementary search result subset as Supplement the search information, determine the supplementary search result set as the search result set, and continue to perform the search step.
18. The apparatus according to claim 17, wherein the supplementary search unit is further configured to:

    In response to determining that the supplementary search result set satisfies the convergence condition, a target search result set is determined from the supplementary search result set.
19. The apparatus according to claim 17, wherein the selecting unit is further configured to:

    Obtain annotation information of search results in a search result subset, where the annotation information is used to indicate whether the search result is a target search result;

    According to the labeled information, the search results in the search result subset that belong to the target search result are determined.
20. The apparatus according to claim 18, wherein the apparatus further comprises:

    The relevance degree determination unit is configured to determine a relevance degree of the search results in the search result set, where the relevance degree is used to indicate a relevance degree of the search results and the search information.
21. The apparatus according to claim 20, wherein the relevance determination unit is further configured to:

    Determine the similarity between the search results and the search information;

    According to the similarity, the relevance of the search result is determined, wherein the similarity is directly proportional to the relevance.
22. The apparatus according to claim 20, wherein the relevance determination unit is further configured to:

    For the supplementary search results in the supplementary search result set, in response to the supplementary search results existing in the search result set, according to a preset correlation degree improvement algorithm, update the correlation degree of the supplementary search results, and use the updated correlation degree as the supplement The relevance of the search results; in response to the supplementary search results not being present in the search result set, the relevance of the supplementary search results is determined.
23. The apparatus according to claim 22, wherein the relevance determination unit is further configured to:

    Determine the similarity between the supplementary search results and the search information;

    The relevance degree of the supplementary search result is determined according to the similarity degree, wherein the similarity degree is directly proportional to the relevance degree.
24. The apparatus according to claim 22, wherein the supplementary search unit is further configured to:

    According to the order of relevance, from the supplementary search result set, the target number of supplementary search results are selected as the target search result, and the target search result set is obtained.
25. The apparatus according to claim 20, wherein the apparatus further comprises a dividing unit configured to:

    Dividing the search result set into a corresponding number of target search result subsets according to a preset number of target relevance intervals; and

    The selecting unit is further configured to:

    The search results are selected from the target number of search result subsets to obtain the search result subsets.
26. The apparatus according to claim 25, wherein the apparatus further comprises:

    The accuracy determination unit is configured to determine the accuracy of the search result subset in the search result subset of the target number, where the accuracy is used to represent the proportion of the target search results in the search result subset.
27. The apparatus according to claim 26, wherein the dividing unit is further configured to:

    Dividing the supplementary search result set into corresponding target number supplementary search result subsets according to the correlation degree interval; and

    The determining unit is further configured to:

    The supplementary search results are selected from the target number of supplementary search result subsets to obtain the supplementary search result subsets.
28. The apparatus according to claim 27, wherein the accuracy determination unit is further configured to:

    The accuracy of the supplementary search result subset of the target number of supplementary search result sets is determined, where the accuracy is used to represent the proportion of target search results in the supplementary search result subset.
29. The apparatus according to claim 28, wherein the determination unit is further configured to:

    For the supplementary search subset of the target number of supplementary search result subsets, determine the absolute value of the difference between the accuracy of the search result set corresponding to the correlation interval in which the supplementary search result subset is located and the preset accuracy threshold. Determining the number of supplementary search results selected from the supplementary search result subset according to the absolute value and the number of supplementary search results included in the supplementary search result subset, wherein the absolute value and the number of supplementary search results selected The number is inversely proportional, and the number of supplementary search results included is proportional to the number of selected supplementary search results.
30. The apparatus according to any one of claims 17 to 29, wherein the number of search results in the search result set and the number of supplementary search results in the supplementary search result subset are the same.
31. The apparatus according to claim 29, wherein the convergence conditions include:

    The difference between the accuracy of the search result subset and the supplementary search result subset corresponding to the target relevance degree interval is less than a preset difference threshold.
32. The apparatus according to claim 17, wherein the convergence conditions include:

    The number of search results included in the difference between the search result set and the supplementary search result set is less than a preset number difference threshold.
33. An electronic device includes:

    One or more processors;

    A storage device on which one or more programs are stored;

    When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-16.
34. A computer-readable medium having stored thereon a computer program, wherein the program, when executed by a processor, implements the method of any one of claims 1-16.

Images