CN109800361A - A kind of method for digging of interest point name, device, electronic equipment and storage medium - Google Patents

Abstract

The embodiment of the invention discloses a kind of method for digging of interest point name, device, electronic equipment and storage mediums.The described method includes: obtaining the search sessions of each user in historical search log；The corresponding interest point name pair of each search sessions is determined according to the search sessions of each user；Wherein, the interest point name of the interest point name centering belongs to the same search sessions；Determine each interest point name to corresponding search characteristics in predetermined search characteristics library；According to the corresponding interest point name pair of each search sessions and each interest point name to corresponding search characteristics, the interest point name of each interest point name centering is excavated.The digging efficiency of interest point name not only can be improved, it can also be ensured that the accuracy rate of interest point name.

Description

A method, device, electronic device and storage medium for mining the name of a point of interest

technical field

Embodiments of the present invention relate to the field of Internet technologies, and in particular, to a method, device, electronic device, and storage medium for mining a name of a point of interest.

Background technique

There are a large number of positioning points in the electronic map, such as restaurants, hotels, scenic spots, toll stations, etc. marked on the map. These positioning points are the points of interest (POI) that users may query or want to reach. .

Under normal circumstances, the same POI will have multiple different names. For example, a POI can have two names, namely: "Lama Temple" and "Lama Temple"; in this case, "Lama Temple" can be used as the The original name of the POI; the original name is the name that already exists in the existing electronic map; "Lama Temple" can be used as the candidate name of the POI; the candidate name does not exist in the existing electronic map name. In the existing electronic map database, it is likely that not all the names of a point of interest are contained. For example, for the point of interest with the original name of "Lama Temple", only the original name of "Lama Temple" exists in the database in the electronic map, but the candidate name of "Lama Temple" does not exist. Then, if a user searches for a point of interest named "Lama Temple" in the electronic map, since the candidate name "Lama Temple" does not exist in the database in the electronic map, the user cannot find this point of interest.

In order to improve the user's search experience, the candidate names of each point of interest need to be mined, and then the mined candidate names are added to the electronic map database. In the existing mining methods of POI names, the POI names are usually mined manually, which not only has low mining efficiency, but also cannot guarantee the accuracy.

SUMMARY OF THE INVENTION

In view of this, the embodiments of the present invention provide a method, device, electronic device and storage medium for mining POI names, which can not only improve the mining efficiency of POI names, but also ensure the accuracy of POI names.

In a first aspect, an embodiment of the present invention provides a method for mining POI names, the method comprising:

Obtain the search sessions of each user in the historical search log;

Determine the POI name pairs corresponding to each search session according to the search sessions of each user; wherein, the POI names in the POI name pairs belong to the same search session;

Determine the search feature corresponding to each point of interest name pair in the predetermined search feature library;

According to the POI name pairs corresponding to each search session and the search features corresponding to each POI name pair, the POI names in each POI name pair are mined.

In the above embodiment, the acquisition of the search sessions of each user in the historical search log includes:

Obtain all the POI search terms of each user and the search time corresponding to each POI search term in the historical search log;

All POI search terms of each user are divided into corresponding search sessions according to the search time corresponding to each POI search term; wherein, all POI search terms in each search session are from the same user.

In the above-mentioned embodiment, the determining of the name pairs of points of interest corresponding to each search session according to each user's search session includes:

Determine the POI search terms in each search session in each user's search session;

Every two POI search terms in each search session are combined into a point-of-interest name pair, which is used as a point-of-interest name pair corresponding to each search session.

In the above embodiment, determining the search features corresponding to each point of interest name pair in the predetermined search feature library includes:

Determine the search result feature corresponding to each POI name pair in a predetermined search result feature library; wherein, the POI name pair includes: a first POI name and a second POI name;

A user behavior feature corresponding to each POI name pair is determined in a predetermined user behavior feature database; wherein, the user behavior feature includes: the search popularity feature of the first POI name and the search commonality of each POI name pair. present feature.

In the above embodiment, mining the POI names in each POI name pair according to the POI name pairs corresponding to each search session and the corresponding search features of each POI name pair, including:

Input the POI name pair corresponding to each search session and the search feature corresponding to each POI name pair into the pre-trained discriminant model;

The discrimination probability value corresponding to each point of interest name pair is obtained through the discriminant model;

According to the discrimination probability value corresponding to each interest point name pair and the predetermined discrimination threshold, the discrimination result corresponding to each interest point name pair is determined.

In a second aspect, an embodiment of the present invention provides a device for mining a point of interest name, the device includes: an acquisition module, a determination module, and a mining module; wherein,

The obtaining module is used to obtain the search session of each user in the historical search log;

The determining module is used to determine the POI name pairs corresponding to each search session according to the search sessions of each user; wherein, the POI names in the POI name pairs belong to the same search session; in a predetermined search feature database Determine the search feature corresponding to each point of interest name pair in ;

The mining module is configured to mine the POI names in each POI name pair according to the POI name pair corresponding to each search session and the corresponding search feature of each POI name pair.

In the above embodiment, the acquisition module includes: an acquisition sub-module and a division sub-module; wherein,

The acquisition submodule is used to acquire all the POI search terms of each user and the search time corresponding to each POI search term in the historical search log;

The dividing submodule is used to divide all POI search terms of each user into the corresponding search sessions according to the search time corresponding to each POI search term; wherein, all POI search terms in each search session are from the same user.

In the above embodiment, the determining module is specifically configured to determine the POI search terms in each search session in each user's search session; combine every two POI search terms in each search session into a point-of-interest name pair , as the POI name pairs corresponding to each search session.

In the above embodiment, the determining module is specifically configured to determine the search result feature corresponding to each POI name pair in a predetermined search result feature library; wherein, the POI name pair includes: a first POI name and the name of the second point of interest; determine the corresponding user behavior characteristics of each point of interest name in the predetermined user behavior feature library; wherein, the user behavior characteristics include: the search popularity feature of the name of the first point of interest and each Search co-occurrence features for POI name pairs.

In the above embodiment, the mining module is specifically configured to input the name pairs of interest points corresponding to each search session and the search features corresponding to the name pairs of each interest point into the pre-trained discriminant model; Discrimination probability value corresponding to each POI name pair; according to the corresponding discrimination probability value of each POI name pair and a predetermined discrimination threshold, determine the corresponding discrimination result of each POI name pair.

In a third aspect, an embodiment of the present invention provides an electronic device, including:

one or more processors;

memory for storing one or more programs,

When the one or more programs are executed by the one or more processors, the one or more processors implement the method for mining POI names according to any embodiment of the present invention.

In a fourth aspect, an embodiment of the present invention provides a storage medium on which a computer program is stored, and when the program is executed by a processor, implements the method for mining a point of interest name described in any embodiment of the present invention.

The embodiments of the present invention provide a method, device, electronic device and storage medium for mining the name of a point of interest. First, the search sessions of each user are obtained from a historical search log; then the interest corresponding to each search session is determined according to the search sessions of each user. point name pair; determine the search feature corresponding to each point of interest name pair in a predetermined search feature library; then according to the point of interest name pair corresponding to each search session and the corresponding search feature of each point of interest name Mining for POI names in . That is to say, in the technical solution of the present invention, the point-of-interest name pair corresponding to each search session can be determined according to the search session of each user; after the search feature corresponding to each point-of-interest name pair is determined, the corresponding The POI name pairs of , and the corresponding search features of each POI name pair, and the POI names in each POI name pair are mined. However, in the existing mining methods of POI names, the POI names are usually mined manually, which not only has low mining efficiency, but also cannot guarantee the accuracy. Therefore, compared with the prior art, the mining method, device, electronic device and storage medium for POI names proposed in the embodiments of the present invention can not only improve the mining efficiency of POI names, but also ensure the accuracy of POI names; In addition, the technical solutions of the embodiments of the present invention are simple and convenient to implement, easy to popularize, and have wider application range.

Description of drawings

1 is a schematic flowchart of a method for mining POI names according to Embodiment 1 of the present invention;

2 is a schematic flowchart of a method for mining POI names according to Embodiment 2 of the present invention;

3 is a schematic flowchart of a method for mining POI names according to Embodiment 3 of the present invention;

FIG. 4 is a first structural schematic diagram of the excavation device for POI names according to Embodiment 4 of the present invention;

FIG. 5 is a second structural schematic diagram of the excavation device for POI names according to Embodiment 4 of the present invention;

FIG. 6 is a schematic structural diagram of an electronic device according to Embodiment 5 of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention. In addition, it should be noted that, for the convenience of description, the drawings only show some but not all of the contents related to the present invention.

Example 1

1 is a schematic flowchart of a method for mining POI names according to Embodiment 1 of the present invention. The method may be executed by a POI name mining device or electronic device, and the device or electronic device may be implemented by software and/or hardware. Implementation, the apparatus or electronic device can be integrated in any smart device with network communication function. As shown in Figure 1, the mining method of POI names can include the following steps:

S101. Acquire search sessions of each user from a historical search log.

In a specific embodiment of the present invention, the electronic device may acquire the search sessions of each user in the historical search log. Specifically, the electronic device can first obtain all the POI search terms of each user and the search time corresponding to each POI search term in the historical search log; then, according to the search time corresponding to each POI search term, all POI search terms of each user are divided into in the corresponding search session; wherein, all POI search terms in each search session are from the same user. For example, suppose a user's historical search log includes N POI search terms, namely: POI search term 1, POI search term 2, ..., POI search term N; where N is a natural number greater than 1; where POI search term The search time corresponding to word 1 is search time 1, the search time corresponding to POI search word 2 is search time 2, ..., and the search time corresponding to POI search word N is search time N. In a specific embodiment of the present invention, if the difference between the search times corresponding to two consecutive POI search terms is less than or equal to a preset threshold, it means that the two consecutive POI search terms belong to the same search session; if the two consecutive POI search terms belong to the same search session; The difference between the search times corresponding to the POI search terms is greater than the preset threshold, indicating that the two related POI search terms do not belong to the same search session.

S102. Determine, according to the search sessions of each user, a point-of-interest name pair corresponding to each search session; wherein, the point-of-interest names in the point-of-interest name pair belong to the same search session.

In a specific embodiment of the present invention, the electronic device may determine the POI name pairs corresponding to each search session according to the search sessions of each user; wherein the POI names in the POI name pairs belong to the same search session. For example, suppose a user's historical search log may include N POI search terms, namely: POI search term 1, POI search term 2, ..., POI search term N; where N is a natural number greater than 1; where POI The search time corresponding to the search term 1 is the search time 1, the search time corresponding to the POI search term 2 is the search time 2, ..., and the search time corresponding to the POI search term N is the search time N. Assuming that the search time 1 corresponding to the POI search term 1 and the search time 2 corresponding to the search term 2 are less than the preset threshold, it means that the POI search term 1 and the POI search term 2 belong to the same search session. The POI search term 2 is determined as a point-of-interest name pair corresponding to the search session. It is further assumed that the search time 2 corresponding to the POI search term 2 and the search time 3 corresponding to the search term 3 are less than the preset threshold, indicating that the POI search term 2 and the POI search term 3 also belong to the same search session. Therefore, the POI search term can also be searched. Word 2 and POI search word 3 are determined as the point-of-interest name pair corresponding to the search session.

S103: Determine the search feature corresponding to each point of interest name pair in a predetermined search feature library.

In a specific embodiment of the present invention, the electronic device may determine a search feature corresponding to each point of interest name pair in a predetermined search feature library. Specifically, the electronic device may determine the search result feature corresponding to each POI name pair in the predetermined search result feature library; wherein, the POI name pair includes: the first POI name and the second POI name; in the present invention In the specific embodiment of , the name of the point of interest is required to be in the middle, and the second POI search term can correspond to a POI in the electronic map database; the method for determining that the second POI search term can correspond to a POI in the electronic map database is: : When the user searches for the second POI search term, the user can obtain the search result of the second POI search term through the POI search engine, and clicks the POI in the search result of the second POI search term, and the clicked POI is used as The POI corresponding to the second POI search term. In addition, the electronic device may also determine the user behavior characteristics corresponding to each POI name pair in a predetermined user behavior characteristic database; wherein, the user behavior characteristics include: the search popularity characteristics of the first POI name and the corresponding user behavior characteristics of each POI name pair. Search for co-occurrence features.

S104 , mining the POI names in each POI name pair according to the POI name pair corresponding to each search session and the search feature corresponding to each POI name pair.

In a specific embodiment of the present invention, the electronic device may mine the POI names in each POI name pair according to the POI name pairs corresponding to each search session and the search features corresponding to each POI name pair. Specifically, the electronic device can input the name pairs of the points of interest corresponding to each search session and the search features corresponding to the name pairs of the points of interest into the pre-trained discriminant model; through the discriminant model, the corresponding discriminants of the name pairs of the respective points of interest can be obtained Probability value; according to the corresponding discrimination probability value of each interest point name pair and the predetermined discrimination threshold, determine the discrimination result corresponding to each interest point name pair. For example, for a point-of-interest name pair (POI search term 1, POI search term 2), POI search term 2 is the original name of the POI; the original name is a name that already exists in an existing electronic map; POI search term 1 It is the suspected candidate name of the POI; the suspected candidate name is the candidate name to be added to the existing electronic map. In this step, the electronic device may input the POI name pair (POI search term 1, POI search term 2) and the search feature corresponding to the POI name pair into the pre-trained discriminant model; it is assumed that the discriminant model obtains The corresponding discrimination probability value of the POI name pair is 80%; assuming that the predetermined discrimination threshold is 50%; it can be determined that POI search term 1 is a candidate name of POI corresponding to POI search term 2.

The method for mining POI names proposed by the embodiment of the present invention first acquires the search sessions of each user in the historical search log; then determines the POI name pairs corresponding to each search session according to the search sessions of each user; The search feature corresponding to each POI name pair is determined in the database; and then the POI name in each POI name pair is mined according to the POI name pair corresponding to each search session and the corresponding search feature of each POI name pair. That is to say, in the technical solution of the present invention, the point-of-interest name pair corresponding to each search session can be determined according to the search session of each user; after the search feature corresponding to each point-of-interest name pair is determined, the corresponding The POI name pairs of , and the corresponding search features of each POI name pair, and the POI names in each POI name pair are mined. However, in the existing mining methods of POI names, the POI names are usually mined manually, which not only has low mining efficiency, but also cannot guarantee the accuracy. Therefore, compared with the prior art, the mining method for POI names proposed by the embodiments of the present invention can not only improve the mining efficiency of POI names, but also ensure the accuracy of POI names; The implementation of the scheme is simple and convenient, easy to popularize, and has a wider application range.

Embodiment 2

FIG. 2 is a schematic flowchart of a method for mining POI names according to Embodiment 2 of the present invention. As shown in Figure 2, the mining method of POI names can include the following steps:

S201. Acquire all the POI search terms of each user and the search time corresponding to each POI search term from the historical search log.

In a specific embodiment of the present invention, the electronic device may acquire all the POI search terms of each user and the search time corresponding to each POI search term from the historical search log. For example, suppose a user's historical search log may include N POI search terms, namely: POI search term 1, POI search term 2, ..., POI search term N; where N is a natural number greater than 1; where POI The search time corresponding to the search term 1 is the search time 1, the search time corresponding to the POI search term 2 is the search time 2, ..., and the search time corresponding to the POI search term N is the search time N.

S202: Divide all the POI search terms of each user into their corresponding search sessions according to the search time corresponding to each POI search term; wherein, all POI search terms in each search session are from the same user.

In a specific embodiment of the present invention, the electronic device may divide all the POI search terms of each user into their corresponding search sessions according to the search time corresponding to each POI search term; wherein, all POI search terms in each search session come from the same user. Specifically, in a specific embodiment of the present invention, if the difference between the search times corresponding to two consecutive POI search terms is less than or equal to a preset threshold, it means that the two consecutive POI search terms belong to the same search session; if The difference between the search times corresponding to two consecutive POI search terms is greater than the preset threshold, indicating that the two related POI search terms do not belong to the same search session. For example, assuming that the search time 1 corresponding to the POI search term 1 and the search time 2 corresponding to the search term 2 are less than the preset threshold, it means that the POI search term 1 and the POI search term 2 belong to the same search session. Therefore, in this step, The electronic device can divide the POI search term 1 and the POI search term 2 into the search session; it is then assumed that the search time 2 corresponding to the POI search term 2 and the search time 3 corresponding to the search term 3 are less than the preset threshold, indicating that the POI search term 2 and POI search term 3 also belong to the same search session. Therefore, in this step, the electronic device may also divide POI search term 2 and POI search term 3 into the search session.

S203. Determine the POI search words in each search session in each user's search session.

In a specific embodiment of the present invention, the electronic device may determine the POI search terms in each search session of each user in each user's search session. For example, suppose a user's search session includes three POI search terms, namely: POI search term q1, POI search term q2, and POI search term q3; in this step, the electronic device can determine the POI search term in the search session Search term q1, POI search term q2, and POI search term q3.

S204. Combine every two POI search terms in each search session into a point-of-interest name pair, which is used as a point-of-interest name pair corresponding to each search session.

In a specific embodiment of the present invention, the electronic device may combine every two POI search terms in each search session into a point-of-interest name pair, as a point-of-interest name pair corresponding to each search session. For example, suppose a user's search session includes three POI search terms, namely: POI search term q1, POI search term q2, and POI search term q3; in this step, the electronic device can search POI search term q1 and POI search term The word q2 is combined into a point-of-interest name pair, which is the point-of-interest name pair corresponding to the search session; the POI search word q2 and the POI search word q3 can also be combined into a point-of-interest name pair, which is used as the point-of-interest name corresponding to the search session. Yes; the POI search term q1 and the POI search term q3 can also be combined into a point-of-interest name pair, which is used as a point-of-interest name pair corresponding to the search session.

S205: Determine the search feature corresponding to each point of interest name pair in a predetermined search feature library.

In a specific embodiment of the present invention, the electronic device may determine a search feature corresponding to each point of interest name pair in a predetermined search feature database. Specifically, the electronic device may determine the search result feature corresponding to each POI name pair in the predetermined search result feature library; wherein, the POI name pair includes: the first POI name and the second POI name; in the present invention In the specific embodiment of , the name of the point of interest is required to be in the middle, and the second POI search term can correspond to a POI in the electronic map database; the method for determining that the second POI search term can correspond to a POI in the electronic map database is: : When the user searches for the second POI search term, the user can obtain the search result of the second POI search term through the POI search engine, and clicks the POI in the search result of the second POI search term, and the clicked POI is used as The POI corresponding to the second POI search term. In addition, the electronic device may also determine the user behavior characteristics corresponding to each POI name pair in a predetermined user behavior characteristic database; wherein, the user behavior characteristics include: the search popularity characteristics of the first POI name and the corresponding user behavior characteristics of each POI name pair. Search for co-occurrence features.

S206 , mining the POI names in each POI name pair according to the POI name pair corresponding to each search session and the search feature corresponding to each POI name pair.

In a specific embodiment of the present invention, the electronic device may mine the POI names in each POI name pair according to the POI name pairs corresponding to each search session and the search features corresponding to each POI name pair. Specifically, the electronic device can input the name pairs of the points of interest corresponding to each search session and the search features corresponding to the name pairs of the points of interest into the pre-trained discriminant model; through the discriminant model, the corresponding discriminants of the name pairs of the respective points of interest can be obtained Probability value; according to the corresponding discrimination probability value of each interest point name pair and the predetermined discrimination threshold, determine the discrimination result corresponding to each interest point name pair. For example, for a point-of-interest name pair (POI search term 1, POI search term 2), POI search term 2 is the original name of the POI; the original name is a name that already exists in an existing electronic map; POI search term 1 It is the suspected candidate name of the POI; the suspected candidate name is the candidate name to be added to the existing electronic map. In this step, the electronic device may input the POI name pair (POI search term 1, POI search term 2) and the search feature corresponding to the POI name pair into the pre-trained discriminant model; it is assumed that the discriminant model obtains The corresponding discrimination probability value of the POI name pair is 80%; assuming that the predetermined discrimination threshold is 50%; it can be determined that POI search term 1 is a candidate name of POI corresponding to POI search term 2.

The method for mining POI names proposed by the embodiment of the present invention first acquires the search sessions of each user in the historical search log; then determines the POI name pairs corresponding to each search session according to the search sessions of each user; The search feature corresponding to each POI name pair is determined in the database; and then the POI name in each POI name pair is mined according to the POI name pair corresponding to each search session and the corresponding search feature of each POI name pair. That is to say, in the technical solution of the present invention, the point-of-interest name pair corresponding to each search session can be determined according to the search session of each user; after the search feature corresponding to each point-of-interest name pair is determined, the corresponding The POI name pairs of , and the corresponding search features of each POI name pair, and the POI names in each POI name pair are mined. However, in the existing mining methods of POI names, the POI names are usually mined manually, which not only has low mining efficiency, but also cannot guarantee the accuracy. Therefore, compared with the prior art, the mining method for POI names proposed by the embodiments of the present invention can not only improve the mining efficiency of POI names, but also ensure the accuracy of POI names; The implementation of the scheme is simple and convenient, easy to popularize, and has a wider application range.

Embodiment 3

FIG. 3 is a schematic flowchart of a method for mining POI names according to Embodiment 3 of the present invention. As shown in Figure 3, the mining method of POI names can include the following steps:

S301. Acquire all the POI search terms of each user and the search time corresponding to each POI search term from the historical search log.

In a specific embodiment of the present invention, the electronic device may acquire all the POI search terms of each user and the search time corresponding to each POI search term from the historical search log. For example, suppose a user's historical search log may include N POI search terms, namely: POI search term 1, POI search term 2, ..., POI search term N; where N is a natural number greater than 1; where POI The search time corresponding to the search term 1 is the search time 1, the search time corresponding to the POI search term 2 is the search time 2, ..., and the search time corresponding to the POI search term N is the search time N.

S302. Divide all the POI search terms of each user into the corresponding search sessions according to the search time corresponding to each POI search term; wherein, all the POI search terms in each search session are from the same user.

In a specific embodiment of the present invention, the electronic device may divide all the POI search terms of each user into their corresponding search sessions according to the search time corresponding to each POI search term; wherein, all POI search terms in each search session come from the same user. Specifically, in a specific embodiment of the present invention, if the difference between the search times corresponding to two consecutive POI search terms is less than or equal to a preset threshold, it means that the two consecutive POI search terms belong to the same search session; if The difference between the search times corresponding to two consecutive POI search terms is greater than the preset threshold, indicating that the two related POI search terms do not belong to the same search session. For example, assuming that the search time 1 corresponding to the POI search term 1 and the search time 2 corresponding to the search term 2 are less than the preset threshold, it means that the POI search term 1 and the POI search term 2 belong to the same search session. Therefore, in this step, The electronic device can divide the POI search term 1 and the POI search term 2 into the search session; it is then assumed that the search time 2 corresponding to the POI search term 2 and the search time 3 corresponding to the search term 3 are less than the preset threshold, indicating that the POI search term 2 and POI search term 3 also belong to the same search session. Therefore, in this step, the electronic device may also divide POI search term 2 and POI search term 3 into the search session.

S303. Determine the POI search words in each search session in each user's search session.

In a specific embodiment of the present invention, the electronic device may determine the POI search terms in each search session of each user in each user's search session. For example, suppose a user's search session includes three POI search terms, namely: POI search term q1, POI search term q2, and POI search term q3; in this step, the electronic device can determine the POI search term in the search session Search term q1, POI search term q2, and POI search term q3.

S304. Combine every two POI search terms in each search session into a point-of-interest name pair, which is used as a point-of-interest name pair corresponding to each search session.

In a specific embodiment of the present invention, the electronic device may combine every two POI search terms in each search session into a point-of-interest name pair, as a point-of-interest name pair corresponding to each search session. For example, suppose a user's search session includes three POI search terms, namely: POI search term q1, POI search term q2, and POI search term q3; in this step, the electronic device can search POI search term q1 and POI search term The word q2 is combined into a point-of-interest name pair, which is the point-of-interest name pair corresponding to the search session; the POI search word q2 and the POI search word q3 can also be combined into a point-of-interest name pair, which is used as the point-of-interest name corresponding to the search session. Yes; the POI search term q1 and the POI search term q3 can also be combined into a point-of-interest name pair, which is used as a point-of-interest name pair corresponding to the search session.

S305: Determine the search result feature corresponding to each POI name pair in a predetermined search result feature library.

In a specific embodiment of the present invention, the electronic device may determine the search result feature corresponding to each point of interest name pair in a predetermined search result feature library. The retrieval results of general search engines are very effective in terms of relevance. For POI name pairs (q _i , q _j ), if q _i is a candidate name of q _j , then the search results of q _i are likely to contain the name represented by q _j . string; on the contrary, the probability of the string represented by q _j appearing in the search results is very low. Therefore, in a specific embodiment of the present invention, the electronic device can use qi as a search term of a general search engine, obtain search results in the general search engine, and then count the number of times _{q j appears} in the search results as the point of interest The name pair corresponds to the corresponding search result feature. Here, a general search engine is a search engine for searching POIs and non-POIs; a POI search engine is a search engine specifically for searching for POIs.

S306. Determine user behavior features corresponding to each POI name pair in a predetermined user behavior feature database; wherein, the user behavior features include: a search popularity feature of the first POI name and a search co-occurrence feature of each POI name pair .

In a specific embodiment of the present invention, the electronic device may determine the user behavior characteristics corresponding to each POI name pair in a predetermined user behavior characteristic database; wherein the user behavior characteristics include: the search popularity characteristics of the first POI name and Search co-occurrence features for each POI name pair. Specifically, for the point-of-interest name pair (qi _i , q _j ), if the search popularity feature of qi is very low _, the probability that qi and _{q j represent} the same POI will be very low. In addition, qi and q _j have appeared in the same search session _, so the probability that qi and _{q j represent} the same POI will be high. In a specific embodiment of the present invention, the search co-occurrence feature of each POI name pair may include: co-occurrence times of each POI name pair, transition probability of each POI name pair, and mutual information of each POI name pair. Specifically, for the POI name pair (qi _i , q _j ), the co-occurrence times of the POI name pair are: in the search sessions of all users within a certain historical period, the search sessions in which qi and _{q j co} -occur The number of times cnt(q _i , q _j ); the transition probability of this POI name pair is: in the search sessions of all users in a certain historical period, the number of co-occurrence search sessions of the POI name pair cnt(q The ratio of _i , q _j ) to the number of times cnt(q _j ) of the search session where q _j appeared; the formula is expressed as The mutual information of this POI name pair represents the interdependence between qi and _{q j ;} it is expressed as Among them, cnt(q _i , q _j ) is the number of search sessions in which qi and _{q j co} -occur in the search sessions of all users in a certain historical period; cnt(q _i ) is all the search sessions in a certain historical period In the user's search sessions, the number of search sessions in which q _i appears; cnt(q _i ) is the number of times in which q _j appears in the search sessions of all users in a certain historical period.

S307 , mining the POI names in each POI name pair according to the POI name pair corresponding to each search session and the search feature corresponding to each POI name pair.

In a specific embodiment of the present invention, the electronic device may mine the POI names in each POI name pair according to the POI name pairs corresponding to each search session and the search features corresponding to each POI name pair. Specifically, the electronic device can input the name pairs of the points of interest corresponding to each search session and the search features corresponding to the name pairs of the points of interest into the pre-trained discriminant model; through the discriminant model, the corresponding discriminants of the name pairs of the respective points of interest can be obtained Probability value; according to the corresponding discrimination probability value of each interest point name pair and the predetermined discrimination threshold, determine the discrimination result corresponding to each interest point name pair. For example, for a point-of-interest name pair (POI search term 1, POI search term 2), POI search term 2 is the original name of the POI; the original name is a name that already exists in an existing electronic map; POI search term 1 It is the suspected candidate name of the POI; the suspected candidate name is the candidate name to be added to the existing electronic map. In this step, the electronic device may input the POI name pair (POI search term 1, POI search term 2) and the search feature corresponding to the POI name pair into the pre-trained discriminant model; it is assumed that the discriminant model obtains The corresponding discrimination probability value of the POI name pair is 80%; assuming that the predetermined discrimination threshold is 50%, it can be determined that the POI search term 1 is a candidate name of the POI search term.

The method for mining POI names proposed by the embodiment of the present invention first acquires the search sessions of each user in the historical search log; then determines the POI name pairs corresponding to each search session according to the search sessions of each user; The search feature corresponding to each POI name pair is determined in the database; and then the POI name in each POI name pair is mined according to the POI name pair corresponding to each search session and the corresponding search feature of each POI name pair. That is to say, in the technical solution of the present invention, the point-of-interest name pair corresponding to each search session can be determined according to the search session of each user; after the search feature corresponding to each point-of-interest name pair is determined, the corresponding The POI name pairs of , and the corresponding search features of each POI name pair, and the POI names in each POI name pair are mined. However, in the existing mining methods of POI names, the POI names are usually mined manually, which not only has low mining efficiency, but also cannot guarantee the accuracy. Therefore, compared with the prior art, the mining method for POI names proposed by the embodiments of the present invention can not only improve the mining efficiency of POI names, but also ensure the accuracy of POI names; The implementation of the scheme is simple and convenient, easy to popularize, and has a wider application range.

Embodiment 4

FIG. 4 is a schematic diagram of a first structure of an excavating device for a point of interest name provided in Embodiment 4 of the present invention. As shown in FIG. 4 , the device for mining POI names according to the embodiment of the present invention may include: an acquisition module 401, a determination module 402, and a mining module 403; wherein,

The obtaining module 401 is used to obtain the search session of each user in the historical search log;

The determining module 402 is configured to determine the POI name pairs corresponding to each search session according to the search sessions of each user; wherein, the POI names in the POI name pairs belong to the same search session; Determine the search feature corresponding to each POI name pair in the library;

The mining module 403 is configured to mine the POI names in each POI name pair according to the POI name pair corresponding to each search session and the search feature corresponding to each POI name pair.

FIG. 5 is a schematic diagram of the second structure of the excavation device for POI names according to Embodiment 4 of the present invention. As shown in FIG. 5, the acquisition module 401 includes: an acquisition sub-module 4011 and a division sub-module 4012; wherein,

The acquisition submodule 4011 is used to acquire all the POI search terms of each user and the search time corresponding to each POI search term in the historical search log;

The dividing submodule 4012 is used to divide all the POI search terms of each user into their corresponding search sessions according to the search time corresponding to each POI search term; wherein, all POI search terms in each search session are from the same user .

Further, the determining module 402 is specifically configured to determine the POI search terms in each search session in each user's search session; combine every two POI search terms in each search session into a point-of-interest name pair, as Point-of-interest name pairs corresponding to each search session.

Further, the determining module 402 is specifically configured to determine the search result feature corresponding to each POI name pair in a predetermined search result feature library; wherein, the POI name pair includes: the first POI name and the first POI name pair. Two point of interest names; determine the user behavior characteristics corresponding to each point of interest name pair in a predetermined user behavior feature database; wherein, the user behavior characteristics include: the search popularity feature of the name of the first point of interest and each point of interest Search co-occurrence features for name pairs.

Further, the mining module 403 is specifically configured to input the name pairs of interest points corresponding to each search session and the search features corresponding to the name pairs of each interest point into the pre-trained discriminant model; obtain each interest point through the discriminant model. The point name pair corresponds to the discrimination probability value; the discrimination result corresponding to each interest point name pair is determined according to the discrimination probability value corresponding to each interest point name pair and the predetermined discrimination threshold.

The above-mentioned mining device for the name of the point of interest can execute the method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method. For technical details not described in detail in this embodiment, reference may be made to the method for mining POI names provided by any embodiment of the present invention.

Embodiment 5

FIG. 6 is a schematic structural diagram of an electronic device according to Embodiment 5 of the present invention. Figure 6 shows a block diagram of an exemplary electronic device suitable for use in implementing embodiments of the present invention. The electronic device 12 shown in FIG. 6 is only an example, and should not impose any limitations on the functions and scope of use of the embodiments of the present invention.

As shown in FIG. 6, the electronic device 12 takes the form of a general-purpose computing device. Components of electronic device 12 may include, but are not limited to, one or more processors or processing units 16 , system memory 28 , and a bus 18 connecting various system components including system memory 28 and processing unit 16 .

Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of a variety of bus structures. By way of example, these architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect ( PCI) bus.

Electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 12, including both volatile and non-volatile media, removable and non-removable media.

System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and/or cache memory 32 . Electronic device 12 may further include other removable/non-removable, volatile/non-volatile computer system storage media. For example only, storage system 34 may be used to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard drive"). Although not shown in Figure 6, a disk drive may be provided for reading and writing to removable non-volatile magnetic disks (eg "floppy disks"), as well as removable non-volatile optical disks (eg CD-ROM, DVD-ROM) or other optical media) to read and write optical drives. In these cases, each drive may be connected to bus 18 through one or more data media interfaces. Memory 28 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of various embodiments of the present invention.

A program/utility 40 having a set (at least one) of program modules 42, which may be stored, for example, in memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data , each or some combination of these examples may include an implementation of a network environment. Program modules 42 generally perform the functions and/or methods of the described embodiments of the present invention.

The electronic device 12 may also communicate with one or more external devices 14 (eg, a keyboard, pointing device, display 24, etc.), with one or more devices that enable a user to interact with the electronic device 12, and/or with Any device (eg, network card, modem, etc.) that enables the electronic device 12 to communicate with one or more other computing devices. Such communication may take place through input/output (I/O) interface 22 . Also, the electronic device 12 may communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network such as the Internet) through a network adapter 20 . As shown, network adapter 20 communicates with other modules of electronic device 12 via bus 18 . It should be appreciated that, although not shown in FIG. 6, other hardware and/or software modules may be used in conjunction with electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tapes drives and data backup storage systems, etc.

The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28, for example, implementing the method for mining the name of the point of interest provided by the embodiment of the present invention.

Embodiment 6

Embodiment 6 of the present invention provides a computer storage medium.

The computer-readable storage medium of the embodiments of the present invention may adopt any combination of one or more computer-readable mediums. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples (a non-exhaustive list) of computer readable storage media include: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), Erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

A computer-readable signal medium may include a propagated data signal in baseband or as part of a carrier wave, with computer-readable program code embodied thereon. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device .

Program code embodied on a computer readable medium may be transmitted using any suitable medium, including - but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations of the present invention may be written in one or more programming languages, including object-oriented programming languages—such as Java, Smalltalk, C++, but also conventional Procedural programming language - such as the "C" language or similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through Internet connection).

Note that the above are only preferred embodiments of the present invention and applied technical principles. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and can also include more other equivalent embodiments without departing from the concept of the present invention. The scope is determined by the scope of the appended claims.

Claims (12)

1. A method for mining a point of interest name, the method comprising:

acquiring search sessions of all users from a historical search log;

determining interest point name pairs corresponding to the search sessions according to the search sessions of the users; wherein the interest point names in the interest point name pair belong to the same search session;

determining search features corresponding to the name pairs of the interest points in a predetermined search feature library;

and mining the interest point names in the interest point name pairs according to the interest point name pairs corresponding to the search sessions and the search characteristics corresponding to the interest point name pairs.

2. The method of claim 1, wherein obtaining search sessions of respective users in the historical search log comprises:

all POI search words of all the users and the search time corresponding to each POI search word are obtained from the historical search log;

dividing all POI search words of each user into corresponding search sessions according to the search time corresponding to each POI search word; wherein all POI search terms in each search session are from the same user.

3. The method of claim 1, wherein determining the interest point name pair corresponding to each search session according to the search session of each user comprises:

determining POI search words in each search session in the search session of each user;

and combining every two POI search words in each search session into an interest point name pair as the interest point name pair corresponding to each search session.

4. The method of claim 1, wherein determining the search feature corresponding to each interest point name pair in a predetermined search feature library comprises:

determining search result characteristics corresponding to the name pairs of the interest points in a predetermined search result characteristic library; wherein the interest point name pair comprises: a first point of interest name and a second point of interest name;

determining user behavior characteristics corresponding to the name pairs of the interest points in a predetermined user behavior characteristic library; wherein the user behavior characteristics include: the search heat characteristic of the first interest point name and the search co-occurrence characteristic of each interest point name pair.

5. The method of claim 1, wherein mining the interest point names in each pair of interest point names according to the pair of interest point names corresponding to each search session and the search features corresponding to each pair of interest point names comprises:

inputting the interest point name pairs corresponding to the search sessions and the search features corresponding to the interest point name pairs into a pre-trained discrimination model;

obtaining the discrimination probability value corresponding to each interest point name pair through the discrimination model;

and determining the discrimination result corresponding to each interest point name pair according to the discrimination probability value corresponding to each interest point name pair and a predetermined discrimination threshold value.

6. An apparatus for mining a point of interest name, the apparatus comprising: the system comprises an acquisition module, a determination module and a mining module; wherein,

the acquisition module is used for acquiring the search conversation of each user in the historical search log;

the determining module is used for determining interest point name pairs corresponding to the search sessions according to the search sessions of the users; wherein the interest point names in the interest point name pair belong to the same search session; determining search features corresponding to the name pairs of the interest points in a predetermined search feature library;

and the mining module is used for mining the interest point names in the interest point name pairs according to the interest point name pairs corresponding to the search sessions and the search characteristics corresponding to the interest point name pairs.

7. The apparatus of claim 6, wherein the obtaining module comprises: acquiring a submodule and a dividing submodule; wherein,

the acquisition sub-module is used for acquiring all POI search words of each user and search time corresponding to each POI search word in the historical search log;

the dividing submodule is used for dividing all POI search words of each user into corresponding search sessions according to the search time corresponding to each POI search word; wherein all POI search terms in each search session are from the same user.

8. The apparatus of claim 6, wherein:

the determining module is specifically configured to determine, in the search session of each user, a POI search term in each search session; and combining every two POI search words in each search session into an interest point name pair as the interest point name pair corresponding to each search session.

9. The apparatus of claim 6, wherein:

the determining module is specifically configured to determine search result features corresponding to the name pairs of the interest points in a predetermined search result feature library; wherein the interest point name pair comprises: a first point of interest name and a second point of interest name; determining user behavior characteristics corresponding to the name pairs of the interest points in a predetermined user behavior characteristic library; wherein the user behavior characteristics include: the search heat characteristic of the first interest point name and the search co-occurrence characteristic of each interest point name pair.

10. The apparatus of claim 6, wherein:

the mining module is specifically used for inputting the interest point name pairs corresponding to the search sessions and the search features corresponding to the interest point name pairs into a pre-trained discrimination model; obtaining the discrimination probability value corresponding to each interest point name pair through the discrimination model; and determining the discrimination result corresponding to each interest point name pair according to the discrimination probability value corresponding to each interest point name pair and a predetermined discrimination threshold value.

11. An electronic device, comprising:

one or more processors;

a memory for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the method of mining point of interest names according to any one of claims 1 to 5.

12. A storage medium on which a computer program is stored which, when being executed by a processor, implements a method of mining a point of interest name as claimed in any one of claims 1 to 5.