CN111460325B

CN111460325B - POI searching method, device and equipment

Info

Publication number: CN111460325B
Application number: CN201910058573.2A
Authority: CN
Inventors: 罗毅
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2019-01-22
Filing date: 2019-01-22
Publication date: 2023-06-27
Anticipated expiration: 2039-01-22
Also published as: CN111460325A

Abstract

The invention provides a POI searching method, device and equipment, wherein the method comprises the following steps: dividing the name of the POI to be searched according to the word, and searching candidate POIs with the same word as the POI to be searched in the electronic map database according to the dividing result and the coordinates of the POI to be searched to obtain a candidate POI list, wherein the distance between the candidate POI and the POI to be searched does not exceed a preset searching range; and determining target search results from the candidate POI list according to the similarity between the POI to be searched and each candidate POI. According to the technical scheme provided by the embodiment of the invention, when searching candidate POIs, the word-based segmentation result is adopted for searching, so that the problems of word segmentation errors, inconsistent word segmentation and search omission caused by abbreviations in POI names can be avoided, and the accuracy of the search result can be improved.

Description

POI searching method, device and equipment

Technical Field

The present invention relates to the field of electronic map technologies, and in particular, to a method, an apparatus, and a device for searching POI.

Background

The electronic map can provide services such as navigation and path planning for people, and is becoming one of the indispensable application software in daily life. The interest points (Point of Interest, POIs) are a type of basic data of the electronic map, and comprise attribute information such as names, addresses, coordinates and the like, wherein the attribute information can be buildings, shops, parking lots, bus stops and the like, and a user can search the POIs through the electronic map, so that assistance is provided for traveling.

The peripheral search is a relatively critical service in the electronic map service, and when the peripheral search of the appointed POI is carried out, the search of the similar POI is required. At present, a relatively common similar POI searching method is a similar POI searching method based on POI name word segmentation, the method firstly carries out word segmentation on POIs to be searched to form a plurality of sub-words, then searches other POIs with the same sub-words on the periphery according to word segmentation results, and returns a POI with high weight accumulation sum (namely high similarity) according to the weight accumulation sum of the POIs to be searched and the same sub-words in the searched POIs.

However, the similar POI searching method based on the POI name word segmentation relies on word segmentation granularity of word segmentation standards and consistency of word segmentation, and similar POIs cannot be searched through word segmentation results if word segmentation errors or word segmentation inconsistencies exist; moreover, in the case of abbreviations in the POI names, search omission is easy to occur, so that the final search result is inaccurate.

Disclosure of Invention

In view of the above, the present invention provides a method, apparatus and device for searching POI, which are used for improving the accuracy of the search result.

In order to achieve the above object, in a first aspect, an embodiment of the present invention provides a POI searching method, including:

Dividing the name of the POI to be searched according to the word, and searching candidate POIs with the same word as the POI to be searched in the electronic map database according to the dividing result and the coordinates of the POI to be searched to obtain a candidate POI list, wherein the distance between the candidate POI and the POI to be searched does not exceed a preset searching range;

and determining target search results from the candidate POI list according to the similarity between the POI to be searched and each candidate POI.

In a second aspect, an embodiment of the present invention provides a POI search device, including:

the searching module is used for segmenting the name of the POI to be searched according to the word, searching candidate POI with the same word as the POI to be searched in the electronic map database according to the segmentation result and the coordinates of the POI to be searched, and obtaining a candidate POI list, wherein the distance between the candidate POI and the POI to be searched does not exceed a preset searching range;

and the determining module is used for determining target search results from the candidate POI list according to the similarity between the POI to be searched and each candidate POI.

In a third aspect, an embodiment of the present invention provides a POI search apparatus, including: a memory and a processor, the memory for storing a computer program; the processor is configured to perform the method of the first aspect described above when the computer program is invoked.

In a fourth aspect, an embodiment of the present invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method according to the first aspect.

According to the POI searching method, device and equipment provided by the embodiment of the invention, firstly, the names of the POIs to be searched are segmented according to the words, and candidate POIs with the same words as the POIs to be searched in the electronic map database are searched according to the segmentation result and the coordinates of the POIs to be searched, so that a candidate POI list is obtained; and then determining target search results from the candidate POI list according to the similarity between the POI to be searched and each candidate POI. Compared with a similar POI searching method based on POI name word segmentation, the technical scheme provided by the embodiment of the invention adopts the word segmentation result to search when searching candidate POIs, so that the problems of word segmentation errors, inconsistent word segmentation and search omission caused by abbreviations in POI names can be avoided, and the accuracy of the search result can be improved.

Drawings

Fig. 1 is a schematic flow chart of a POI searching method according to an embodiment of the present invention;

fig. 2 is a flow chart of another POI searching method according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of an inverted index structure according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a jiugong lattice according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram of a POI searching device according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a POI searching device according to an embodiment of the present invention.

Detailed Description

Aiming at the technical problem that the search result is inaccurate due to word segmentation errors, inconsistent word segmentation or abbreviations in POI names in the conventional similar POI search method based on POI name segmentation, the embodiment of the invention provides a POI search method, device and equipment, which mainly search candidate POIs with the same characters as the POI to be searched in an electronic map database according to the segmentation result and coordinates of the POI to be searched by segmenting the name of the POI to be searched according to the characters, so as to obtain a candidate POI list; and then determining a target search result from the candidate POI list according to the similarity between the POI to be searched and each candidate POI, so as to avoid word segmentation errors, inconsistent word segmentation and search omission caused by abbreviations in POI names, and improve the accuracy of the search result.

The POI searching method, device and equipment provided by the embodiment of the invention can be applied to surrounding searching scenes, POI redundant data detection scenes and other scenes needing similar POI searching, and the embodiment is not particularly limited to specific application scenes.

Embodiments of the present invention are described below with reference to the accompanying drawings.

Fig. 1 is a flow chart of a POI searching method provided by the embodiment of the present invention, as shown in fig. 1, the method provided by the embodiment of the present invention may include the following steps:

s101, segmenting names of POIs to be searched according to words, and searching candidate POIs with the same words as the POIs to be searched in the electronic map database according to segmentation results and coordinates of the POIs to be searched to obtain a candidate POI list.

Specifically, the POI to be searched can be a POI formed according to search content input by a user and the position of the user, and can also be a POI in an electronic map database; taking a surrounding search scene as an example, the POI to be searched is formed according to search content input by a user and the position of the user, the name of the POI to be searched is the search content input by the user, and the coordinates are the current coordinates of the user.

When the name of the POI to be searched is divided, the characters of the Chinese, japanese and Korean and the like composed of a plurality of bytes can be divided into independent one word, and the characters of the American Standard code for information interchange (American Standard Code for Information Interchange, ASCII) composed of a single byte can be divided into independent one word.

When searching, the candidate POI can be searched in a preset searching range, namely the distance between the candidate POI and the POI to be searched does not exceed the preset searching range. The search range may be selected according to actual needs, for example, may be 3km, and the specific size is not particularly limited in this embodiment.

In addition, when searching candidate POIs, the POIs in the searching range in the electronic map database can be segmented according to words and then matched with segmentation results of the POIs to be searched according to words, and the POIs with the same words as the POIs to be searched are used as candidate POIs. In order to improve the searching efficiency, in this embodiment, the spatial index may be established after the POI in the electronic map database is segmented according to the word, and then candidate POI is searched based on the spatial index, and the specific implementation process may refer to the subsequent embodiment.

S102, determining target search results from the candidate POIs according to the similarity between the POIs to be searched and each candidate POI.

Specifically, after searching candidate POIs, calculating the similarity between the POIs to be searched and each candidate POI by adopting an edit distance or Jaccard similarity algorithm; the similarity between the POI to be searched and each candidate POI can be calculated according to the distance and the name similarity between the POI to be searched and each candidate POI; other similarity algorithms may be used to calculate the similarity between the POIs to be searched and each candidate POI, and may be specifically selected according to actual needs, which is not particularly limited in this embodiment.

After calculating the similarity between the POI to be searched and each candidate POI, the first N candidate POIs with high similarity may be selected as target search results according to needs, or candidate POIs with specific similarity (for example, similarity is greater than a preset threshold) may be selected as target search results, and may be selected according to needs in specific implementation, which is not limited in this embodiment.

According to the POI searching method provided by the embodiment, firstly, names of POIs to be searched are segmented according to words, and candidate POIs with the same words as the POIs to be searched in an electronic map database are searched according to segmentation results and coordinates of the POIs to be searched, so that a candidate POI list is obtained; and then determining target search results from the candidate POI list according to the similarity between the POI to be searched and each candidate POI. Compared with the similar POI searching method based on the POI name word segmentation, the technical scheme provided by the embodiment adopts the word segmentation result to search when searching candidate POIs, so that the problems of word segmentation errors, inconsistent word segmentation and search omission caused by abbreviations in the POI names can be avoided, and the accuracy of the search result can be improved.

Fig. 2 is a flow chart of another POI searching method according to an embodiment of the present invention, which is a further optimization supplement to the embodiment shown in fig. 1, and as shown in fig. 2, the method provided in this embodiment may include the following steps:

s201, performing standardization processing on names of all POIs in the electronic map database.

Considering that different description modes may exist for the names of the POIs, the searched similar POIs are not accurate enough, and in order to reduce similarity deviation caused by the isomerism of characters or semantics in the names of the POIs, in order to improve the accuracy of target search results, in this embodiment, before the names of the POIs are segmented according to the characters, the names of the POIs may be standardized.

In particular implementations, the normalization process may include at least one of the following: changing the name of the corresponding POI belonging to the first category in the preset category set into the standard name corresponding to the first category; performing case, full half angle and/or complex conversion on the name; deleting other punctuation marks without brackets in the name; according to the alias configuration table, replacing the alias word in the name with a standard word corresponding to the alias word; and (3) segmenting the names, and performing administrative division word filtering, chinese digital conversion and/or synonym conversion on the segmented names.

The preset classification set may be determined according to a POI classification table, and the first classification may be any one of preset classifications, which represents a classification to which the currently processed POI belongs. When the preset classification set is determined, some secondary classifications include multiple tertiary classifications, and then the secondary classifications can be added into the preset classification set, and standard names corresponding to all the tertiary classifications belonging to the secondary classifications adopt standard names corresponding to the secondary classifications, wherein the same standard name can be a name corresponding to the secondary classifications or other names. For example: the two-level classification 0101 class comprises a plurality of three-level classifications (010100 class- "gas station", 010101 class- "Chinese petrochemical", 010102 class- "Chinese petroleum" and other three-level classifications), and the name corresponding to the 0101 class is "gas station", so that for POIs belonging to the 0101 class, the standard names can be "gas station" all; similarly, for POIs belonging to class 0718, the standard name may all be "lottery".

When the case conversion is carried out, the upper case can be converted into the lower case, and the lower case can be converted into the upper case; similarly, the conversion of the full-angle and complex forms is similar to the conversion of the case, and the conversion mode can be set according to the actual needs, and the embodiment is not particularly limited thereto.

For alternatives to other nouns, for example: the term "bird nest" is a standard term corresponding to "national stadium", and when the name of the POI includes "bird nest", the term "bird nest" may be replaced by "national stadium".

For filtering administrative division words, for example: the division of administrative division "beijing" may be filtered out by beijing corporation.

In the case of performing the chinese-to-digital conversion, the chinese number may be converted to an arabic number, or the arabic number may be converted to a chinese number, and may be selected as needed, and the embodiment is not particularly limited.

For synonym transformations, for example: the synonym of "information" is "consultation", and when the name of POI contains "information", the "information" can be converted into "consultation".

It should be noted that the foregoing examples are merely illustrative, and are not intended to limit the present invention, and specific implementations may be set forth according to actual situations, and the present embodiment is not limited thereto.

S202, segmenting names of all POIs in the electronic map database according to words, and constructing a spatial index of each POI according to the segmentation result and the coordinates of each POI.

In this embodiment, after the names of the POIs in the electronic map database are standardized, the names may be segmented according to words, and then a spatial index of the POIs may be constructed according to the segmentation result and the coordinates of the POIs according to words, so as to improve the searching efficiency of the POIs.

Specifically, when establishing the spatial index, one type of spatial index can be selected from the common spatial data structures; as an optional implementation manner, in this embodiment, the established spatial index is specifically an inverted index, so as to improve the searching efficiency.

In specific implementation, the coordinates of each POI in the electronic map database can be converted into first coordinates with preset precision according to the search range; then, for each POI in the electronic map database, generating a key value by each word in the segmentation result of the POI and the first coordinate of the POI respectively; and establishing an inverted index for the generated key value, wherein the inverted index comprises a POI list corresponding to the key value.

In this embodiment, the preset precision may be determined according to the search range, and when coordinate transformation is performed, the coordinates of the POI may be subjected to geohash encoding, so as to generate a first coordinate, that is, the first coordinate is a geohash value, where the number of bits of the geohash value corresponds to the preset precision; the coordinates of the POI can be amplified according to a preset scale to generate a first coordinate, wherein the preset scale corresponds to the preset precision; for example, the current coordinate has a scale of 10000:1, the search range is 3km, the preset scale may be 1000:1, and when the coordinate conversion is performed, the coordinate of the POI is multiplied by 10 (10000/1000), and the first coordinate of the POI is obtained by rounding. Compared with a geohash value, the coordinate conversion mode of generating the first coordinate by amplifying according to the preset scale can be used for randomly adjusting the amplifying scale and the searching step length according to the requirement, so that POI searching in different ranges can be flexibly performed.

After the conversion of the coordinates, for each POI in the electronic map database, each word in the segmentation result of the POI can be respectively generated into a key value with the first coordinates of the POI. In a specific implementation, for POIs in the electronic map database, the standardization process, word segmentation, coordinate conversion and key value generation may be sequentially performed one by one, or after the standardization process is uniformly performed on all POIs, word segmentation, coordinate conversion and key value generation may be uniformly performed, and the specific implementation manner is not particularly limited. For the case that a POI contains a plurality of different names, a key value can be generated after each name of the POI is segmented according to words.

The POI list corresponding to the key value in the inverted index may specifically store an Identity (ID) of each POI included in the POI list, and may also store other attribute information of the POI. The IDs in the POI list may be ordered, for example, in order from small to large.

For ease of understanding, the following illustrates the inverted index established. FIG. 3 is a schematic diagram of an inverted index structure provided in an embodiment of the present invention, as shown in FIG. 3, the coordinates of POIs with ID1, ID2 and ID3 are (X1, Y1) after coordinate transformation, wherein the name of the POI with ID1 is abc, the name of the POI with ID2 is bcd, and the name of the POI with ID1 is acid, and then key values (X1, Y1, a), (X1, Y1, b), (X1, Y1, c) and (X1, Y1, d) can be generated; in the inverted index correspondingly established, the POI list corresponding to the key value (X1, Y1, a) is { id1, id3}, the POI list corresponding to the key value (X1, Y1, b) is { id1, id2}, the POI list corresponding to the key value (X1, Y1, c) is { id1, id2, id3}, and the POI list corresponding to the key value (X1, Y1, d) is { id2, id3}.

S203, performing standardization processing on the names of the POIs to be searched.

After the spatial index is constructed, candidate POIs of the POIs to be searched can be searched based on the spatial index. Similar to the process of constructing the spatial index, before searching, the name of the POI to be searched may be standardized, and the specific processing procedure is similar to the process of performing the standardization processing on the name of the POI in the electronic map database in S201, which is not described herein.

And S204, searching candidate POIs with the same words as the POIs to be searched in the spatial index according to the segmentation result and the coordinates of the POIs to be searched.

After the spatial index is built, correspondingly, when searching candidate POIs with the same words as the POIs to be searched in the electronic map database according to the segmentation result and the coordinates of the POIs to be searched, specifically, the candidate POIs with the same words as the POIs to be searched can be searched in the spatial index according to the segmentation result and the coordinates of the POIs to be searched.

In the specific searching process, the coordinates of the POI to be searched can be converted into second coordinates with preset precision; then generating a search key value by each word in the segmentation result of the POI to be searched and each coordinate in the coordinate set respectively; and finally searching the POI list corresponding to each search key value in the inverted index, and determining the searched POI list corresponding to each search key value as a candidate POI list. Wherein the coordinate set includes: the second coordinate and a third coordinate with the distance not exceeding the searching range are the same as the second coordinate in precision; the candidate POIs are POIs in the POI list corresponding to the search key value.

Specifically, when the coordinates of the POI to be searched are subjected to coordinate conversion, the coordinate conversion mode corresponds to the coordinate conversion mode adopted when the spatial index is constructed, namely, if the coordinate conversion mode based on geohash coding is adopted when the spatial index is constructed, the coordinates of the POI to be searched are subjected to geohash coding, and a second coordinate is generated; if a coordinate conversion mode based on a preset scale is adopted when the spatial index is constructed, the coordinates of the POI to be searched are amplified according to the preset scale, and a second coordinate is generated; the specific coordinate conversion process may refer to the corresponding description in step S202, which is not repeated here. Similarly, the search key is generated in the same manner as the key is generated in step S202.

When searching candidate POIs in the inverted index, for the case of generating a first coordinate and a second coordinate by adopting a coordinate conversion mode based on a geohash code, the first coordinate, the second coordinate and a third coordinate in a coordinate set are geohash values, and when determining the third coordinate, the coordinate with the same front m bits as the second coordinate can be searched, and each searched coordinate is used as the third coordinate; then, a search key value is generated based on the third coordinate, and a POI list corresponding to each search key value is searched in the inverted index, wherein the value of m is determined according to the search range and the number of bits of the geohash value, for example: the geohash value is 7 bits, the search range is 3km, and m can be 6.

For the situation that a first coordinate and a second coordinate are generated by adopting a coordinate conversion mode based on a preset scale, when a third coordinate is determined, a multi-grid can be established by expanding the preset step length outwards by taking the second coordinate as the center, and the coordinates corresponding to other grids in the multi-grid except the grid where the second coordinate is located are all used as the third coordinate, namely the third coordinate comprises the coordinates, the peripheral distance of the second coordinate, of which the second coordinate is not more than the preset step length; after determining the third coordinate, a search key value may be generated based on the third coordinate, and the inverted index may be searched for a POI list corresponding to each search key value.

Taking a preset step length as 1 as an example, a nine-square grid can be established, and fig. 4 is a schematic structural diagram of the nine-square grid provided by the embodiment of the invention, as shown in fig. 4, the second coordinate is (x, y), and coordinates corresponding to other 8 grids around the second coordinate in the nine-square grid are (x-1, y-1), (x-1, y), (x-1, y+1), (x, y+1), (x+1, y-1) and (x, y-1), respectively; correspondingly, for the POIs to be searched containing n words, n×9 search key values can be generated, and the searched candidate POI list comprises a POI list corresponding to the n×9 search key values.

The preset step length can be specifically determined according to the search range, and when the search range is larger, the value of the preset step length can be increased, for example, the nine grids are continuously expanded outwards to form 16 grids so as to cover the search range.

S205, determining target search results from the candidate POIs according to the similarity between the POIs to be searched and each candidate POI.

In order to improve the calculation efficiency, in this embodiment, the similarity between the POI to be searched and each candidate POI is calculated according to the coordinate distance and the name similarity between the POI to be searched and each candidate POI, wherein the more the number of the same words in the names of the candidate POI and the POI to be searched is considered, the greater the similarity degree is, and in this embodiment, a similarity calculation method based on the number of common words is adopted when calculating the name similarity, so as to improve the calculation efficiency.

When the method is concretely implemented, when a target search result is determined, firstly, carrying out multipath merging on the POI list corresponding to each search key value in the candidate POI list according to the ID in the POI list corresponding to each search key value in the candidate POI list; then, counting the occurrence times of the IDs of the candidate POIs for each candidate POI in the multi-path merging result, and determining the occurrence times of the IDs of the candidate POIs as the number of common words between the POIs to be searched and the candidate POIs; calculating the similarity between the POI to be searched and each candidate POI according to the number of common words and the distance between the POI to be searched and each candidate POI; and finally, determining candidate POIs with the similarity meeting the preset condition as target search results.

Because the POI lists corresponding to different search key values contain the same ID, the more the number of times each ID appears in the candidate POI list, the more the number of common names of the candidate POI corresponding to the ID and the POI to be searched is; and merging POI lists corresponding to the search key values, wherein the obtained result can be used for counting the same word number. Based on the method, when the spatial index is constructed in advance, ID numbers in the established POI list are orderly arranged, then when the multi-path merging is specifically carried out, the method can be realized by adopting a septicator tree algorithm, the POI list corresponding to a plurality of search key values is constructed into a septicator tree, and then a multi-path merging result is returned. Of course, when multi-path merging is performed, other algorithms may be adopted, and in this embodiment, a loser tree algorithm is preferably adopted to improve the sorting efficiency.

After the multi-path merging result is obtained, counting the occurrence times of the IDs of the candidate POIs for each candidate POI, wherein the occurrence times are the number of common words between the POIs to be searched and the candidate POIs; after counting the number of the public words, the similarity of the names between the POI to be searched and the candidate POI can be determined based on the number of the public words, and then the similarity calculation between the POI to be searched and the candidate POI is carried out.

Considering that the larger the ratio of the number of common words in the name is, the larger the similarity between the POI to be searched and the candidate POI is, namely the ratio of the common words can be used for measuring the similarity of the names. In order to improve the similarity calculation efficiency, in this embodiment, before calculating the similarity between the POI to be searched and each candidate POI, the common word ratio corresponding to each candidate POI may be determined according to the number of words included in the name of the POI to be searched, the number of words included in the name of each candidate POI, and the number of common words between the POI to be searched and each candidate POI; and deleting the candidate POIs with the public word ratio smaller than the preset ratio, namely filtering out the candidate POIs with low similarity, and then carrying out similarity calculation between the filtered candidate POIs and the POIs to be searched.

In specific implementation, the common word ratio corresponding to each candidate POI can be determined according to the formula (1);

R _i ＝M _i /min(A，B _i ) (1)

wherein R is _i Representing the public word ratio corresponding to the ith candidate POI, wherein i is more than or equal to 1 and less than or equal to the total number of candidate POIs of the public word ratio to be calculated (namely the total number of candidate POIs in the candidate POI list); m is M _i Representing the number of common words between the POI to be searched and the ith candidate POI, A represents the number of words contained in the name of the POI to be searched, B _i The number of words contained in the name representing the i candidate POI.

For example: the name of the POI to be searched is Fang Heng International, the name of the candidate POI is Square Heng Hotel, and the public word ratio is 2/4=0.5; also for example: the name of the POI to be searched is "Square Hengniao hotel", the name of the candidate POI is "Square Hengbao hotel", and the public word ratio is 4/4=1.0; for another example: the name of the POI to be searched is "Fang Heng hairdressing", the name of the candidate POI is "Fang Hengjing cut, and the common word ratio is 2/6=0.33.

After the common word ratio is calculated, candidate POIs with the common word ratio smaller than the preset ratio are filtered, wherein the preset ratio can be set according to the requirement, for example, can be set to 0.5, and the specific size is not particularly limited in the embodiment.

When candidate POIs are filtered and the similarity between the POI to be searched and each candidate POI is calculated, in order to further improve the calculation efficiency, when the number of the candidate POIs after filtering is excessive, that is, the number is greater than a preset threshold value L, the first L candidate POIs with higher public word ratio are selected to perform subsequent similarity calculation, wherein the value of L can be set according to requirements, for example, the value can be 350.

When the similarity calculation is specifically performed, the similarity between the POI to be searched and each candidate POI can be calculated according to a formula (2);

S _j ＝log(D _j /(c+M _j )) (2)

Wherein S is _j Representing the similarity between the POI to be searched and the jth candidate POI, wherein j is more than or equal to 1 and less than or equal to the total number of candidate POIs of which the similarity is to be calculated; d (D) _j Representing the distance between the POI to be searched and the jth candidate POI, M _j Representing POI and jth to be searchedThe number of common words between candidate POIs, c, is a constant greater than 0, used to ensure that the denominator is not 0, which may be, for example, 0.1.

Wherein the total number of candidate POIs for which the similarity is to be calculated is determined according to an actual scheme, for example: after the number of the public words between the POI to be searched and the candidate POI is determined, directly calculating the similarity between the POI to be searched and the candidate POI, wherein the total number of the candidate POIs with the similarity to be calculated is the total number of the candidate POIs in the candidate POI list; if candidate POIs based on the common word ratio are filtered before the similarity calculation is performed, the total number of candidate POIs to be calculated is the total number of candidate POIs remaining after the candidate POIs with the common word ratio smaller than the preset ratio are deleted in the candidate POI list; if candidate POI screening based on a preset threshold L is performed before similarity calculation is performed, the total number of candidate POIs to be calculated in similarity is L.

After calculating the similarity between the POI to be searched and each candidate POI, when determining the target search result, specifically, candidate POIs with similarity greater than the preset similarity may be determined as the target search result, or after the calculated similarity is sorted in the order from high to low, candidate POIs corresponding to the first N similarity may be determined as the target search result, where the value of the preset similarity and the value of N may be set according to the needs, and this embodiment is not particularly limited.

According to the POI searching method, through standardized processing of names of all POIs and POIs to be searched in the electronic map database, similarity deviation caused by isomerism of characters or semantics in the POI names can be reduced, and accuracy of target searching results is improved; by adopting the similarity calculation method based on the public word number to calculate the similarity between the POI to be searched and each candidate POI, the calculation efficiency can be effectively improved.

Based on the same inventive concept, as an implementation of the above method, the embodiment of the present invention provides a POI search device, where the embodiment of the device corresponds to the embodiment of the foregoing method, and for convenience of reading, the embodiment of the present invention does not describe details in the embodiment of the foregoing method one by one, but it should be clear that the device in the embodiment can correspondingly implement all the details in the embodiment of the foregoing method.

Fig. 5 is a schematic structural diagram of a POI searching device provided by an embodiment of the present invention, as shown in fig. 5, where the device provided by the embodiment includes:

the searching module 110 is configured to segment the name of the POI to be searched according to the word, and search candidate POIs having the same word as the POI to be searched in the electronic map database according to the segmentation result and the coordinates of the POI to be searched to obtain a candidate POI list, where the distance between the candidate POI and the POI to be searched does not exceed a preset searching range;

The determining module 120 is configured to determine a target search result from the candidate POI list according to the similarity between the POI to be searched and each candidate POI.

As an alternative implementation manner of the embodiment of the present invention, the apparatus further includes:

the construction module 130 is configured to divide the name of each POI in the electronic map database according to the word before the search module 110 divides the name of the POI to be searched according to the word, and construct a spatial index of the POI according to the division result and the coordinate of each POI according to the word;

the searching module 110 is specifically configured to search the spatial index for candidate POIs having the same word as the POI to be searched according to the segmentation result and the coordinates of the POI to be searched.

As a specific implementation manner of the embodiment of the present invention, the building module 130 may specifically be configured to:

according to the search range, converting the coordinates of each POI in the electronic map database into first coordinates with preset precision;

for each POI in the electronic map database, generating a key value by each word in the segmentation result of the POI and a first coordinate of the POI respectively;

and establishing an inverted index for the generated key value, wherein the inverted index comprises a POI list corresponding to the key value.

As a specific implementation manner of the embodiment of the present invention, the search module 110 is specifically configured to:

Converting the coordinates of the POI to be searched into second coordinates with preset precision;

generating a search key value by each word in the segmentation result of the POI to be searched and each coordinate in a coordinate set, wherein the coordinate set comprises: the second coordinate and a third coordinate with the distance not exceeding the searching range are the same as the second coordinate in precision;

searching the POI list corresponding to each search key value in the inverted index, and determining the searched POI list corresponding to each search key value as a candidate POI list, wherein the candidate POI is the POI in the POI list corresponding to the search key value.

As another specific implementation of an embodiment of the present invention, the building module 130 may specifically be configured to: amplifying the coordinates of each POI in the electronic map database according to a preset scale to generate a first coordinate, wherein the preset scale corresponds to the preset precision;

the search module 110 is specifically configured to: amplifying the coordinates of the POI to be searched according to a preset scale to generate second coordinates;

wherein the third coordinates comprise: the periphery of the second coordinate is not more than the coordinate of the preset step length, and the preset step length is determined according to the searching range.

As an optional implementation manner of the embodiment of the invention, the identity number ID of each POI contained in the POI list is stored in the POI list, and the IDs in the POI list are orderly arranged; the determining module 120 is specifically configured to:

According to IDs in the POI list corresponding to each search key value in the candidate POI list, carrying out multipath merging on the POI list corresponding to each search key value in the candidate POI list;

counting the occurrence times of the IDs of the candidate POIs for each candidate POI in the multi-path merging result, and determining the occurrence times of the IDs of the candidate POIs as the number of common words between the POIs to be searched and the candidate POIs;

calculating the similarity between the POI to be searched and each candidate POI according to the number of common words and the distance between the POI to be searched and each candidate POI;

and determining the candidate POIs with the similarity meeting the preset conditions as target search results.

As an alternative implementation of the embodiment of the present invention, the determining module 120 is further configured to:

before calculating the similarity between the POI to be searched and each candidate POI, determining the corresponding public word ratio of each candidate POI according to the number of words contained in the name of the POI to be searched, the number of words contained in the name of each candidate POI and the number of public words between the POI to be searched and each candidate POI;

candidate POIs whose common word ratio is smaller than a preset ratio are deleted.

As a specific implementation manner of the embodiment of the present invention, the determining module 120 is specifically configured to:

determining a common word ratio corresponding to each candidate POI according to a formula (1);

R _i ＝M _i /min(A，B _i ) (1)

Wherein R is _i Representing the public word ratio corresponding to the i candidate POI, wherein i is more than or equal to 1 and less than or equal to the total number of candidate POIs of which the public word ratio is to be calculated; m is M _i Representing the number of common words between the POI to be searched and the ith candidate POI, A represents the number of words contained in the name of the POI to be searched, B _i The number of words contained in the name representing the i candidate POI.

As another specific implementation manner of the embodiment of the present invention, the determining module 120 is specifically configured to:

calculating the similarity between the POI to be searched and each candidate POI according to the formula (2);

S _j ＝log(D _j /(c+M _j )) (2)

wherein S is _j Representing the similarity between the POI to be searched and the jth candidate POI, wherein j is more than or equal to 1 and less than or equal to the total number of candidate POIs of which the similarity is to be calculated; d (D) _j Representing the distance between the POI to be searched and the jth candidate POI, M _j Representing the number of common words between the POI to be searched and the j candidate POI, c is a constant greater than 0.

the processing module 140 is configured to perform a normalization process on the names of the POIs in the electronic map database before the construction module 130 performs word-wise segmentation on the names of the POIs in the electronic map database;

the processing module 140 is further configured to perform normalization processing on the name of the POI to be searched before the searching module 110 performs word segmentation on the name of the POI to be searched.

Wherein the normalization process may include at least one of:

changing the name of the corresponding POI belonging to the first category in the preset category set into the standard name corresponding to the first category;

performing case, full half angle and/or complex conversion on the name;

deleting other punctuation marks without brackets in the name;

according to the alias configuration table, replacing the alias word in the name with a standard word corresponding to the alias word;

and (3) segmenting the names, and performing administrative division word filtering, chinese digital conversion and/or synonym conversion on the segmented names.

The device provided in this embodiment may perform the above method embodiment, and its implementation principle is similar to that of the technical effect, and will not be described herein again.

Based on the same inventive concept, the embodiment of the invention also provides POI searching equipment. Fig. 6 is a schematic structural diagram of a POI search device provided in an embodiment of the present invention, and as shown in fig. 6, the POI search device provided in the embodiment includes: a memory 210 and a processor 220, the memory 210 for storing a computer program; the processor 220 is configured to perform the method described in the method embodiments above when the computer program is invoked.

The POI searching device provided in this embodiment may execute the above method embodiment, and its implementation principle is similar to that of the technical effect, and will not be described herein again.

The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when being executed by a processor, implements the method described in the above method embodiment.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein.

The processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.

Claims

1. A POI search method, comprising:

according to the search range, converting the coordinates of each POI in the electronic map database into first coordinates with preset precision; for each POI in the electronic map database, generating a key value by each word in the segmentation result of the POI and a first coordinate of the POI respectively; establishing an inverted index for the generated key value, wherein the inverted index comprises a POI list corresponding to the key value;

splitting the name of the POI to be searched according to the word, searching candidate POI with the same word as the POI to be searched in the electronic map database according to the splitting result and the coordinates of the POI to be searched, and obtaining a candidate POI list, wherein the method comprises the following steps: converting the coordinates of the POI to be searched into second coordinates with preset precision; generating a search key value by each word in the segmentation result of the POI to be searched and each coordinate in a coordinate set, searching a POI list corresponding to each search key value in the inverted index, and determining the searched POI list corresponding to each search key value as a candidate POI list, wherein the coordinate set comprises: a second coordinate and a third coordinate of which the distance between the second coordinate and the third coordinate does not exceed the searching range, wherein the third coordinate and the second coordinate have the same precision; the distance between the candidate POI and the POI to be searched does not exceed a preset searching range, and the candidate POI is a POI in a POI list corresponding to the searching key value;

2. The method according to claim 1, wherein the converting the coordinates of each POI in the electronic map database into the first coordinates with the preset accuracy includes:

amplifying the coordinates of each POI in the electronic map database according to a preset scale to generate a first coordinate, wherein the preset scale corresponds to the preset precision;

the converting the coordinate of the POI to be searched into the second coordinate with preset precision includes:

amplifying the coordinates of the POI to be searched according to a preset scale to generate a second coordinate;

wherein the third coordinates include: and the peripheral distance of the second coordinate is not more than the coordinate of a preset step length, and the preset step length is determined according to the searching range.

3. The method according to claim 1, wherein the POI list stores therein the identification numbers IDs of the POIs contained in the POI list, and the IDs in the POI list are arranged in order; the determining a target search result from the candidate POI list according to the similarity between the POI to be searched and each candidate POI comprises:

According to IDs in the POI list corresponding to the search key values in the candidate POI list, carrying out multipath merging on the POI list corresponding to the search key values in the candidate POI list;

4. A method according to claim 3, wherein prior to said calculating the similarity between the POI to be searched and each candidate POI, the method further comprises:

determining a public word ratio corresponding to each candidate POI according to the number of words contained in the name of the POI to be searched, the number of words contained in the name of each candidate POI and the number of public words between the POI to be searched and each candidate POI;

5. The method of claim 4, wherein the determining the common word ratio corresponding to each candidate POI according to the number of words included in the name of the POI to be searched, the number of words included in the name of each candidate POI, and the number of common words between the POI to be searched and each candidate POI comprises:

R _i ＝M _i /min(A，B _i ) (1)

6. A method according to claim 3, wherein said calculating the similarity between the POI to be searched and each candidate POI based on the number of common words and the distance between the POI to be searched and each candidate POI comprises:

calculating the similarity between the POI to be searched and each candidate POI according to a formula (2);

S _j ＝log(D _j /(c+M _j )) (2)

wherein S is _j Representing the phase between the POI to be searched and the jth candidate POISimilarity, j is more than or equal to 1 and less than or equal to the total number of candidate POIs of which the similarity is to be calculated; d (D) _j Representing the distance between the POI to be searched and the jth candidate POI, M _j Representing the number of common words between the POI to be searched and the j candidate POI, c is a constant greater than 0.

7. The method of any of claims 2-6, wherein prior to said word-wise slicing the names of the POIs in the electronic map database, the method further comprises:

Carrying out standardized processing on the names of all POIs in the electronic map database;

before the name of the POI to be searched is segmented according to words, the method further comprises the following steps:

and carrying out standardization processing on the names of the POIs to be searched.

8. The method of claim 7, wherein the normalization process comprises at least one of:

performing case, full half angle and/or complex conversion on the name;

deleting other punctuation marks without brackets in the name;

according to the alias configuration table, replacing the alias word in the name with the standard word corresponding to the alias word;

9. A POI search device, comprising:

The determining module is used for determining target search results from the candidate POI list according to the similarity between the POI to be searched and each candidate POI;

the construction module is used for converting the coordinates of each POI in the electronic map database into first coordinates with preset precision according to the search range; for each POI in the electronic map database, generating a key value by each word in the segmentation result of the POI and a first coordinate of the POI respectively; establishing an inverted index for the generated key value, wherein the inverted index comprises a POI list corresponding to the key value;

the searching module is specifically configured to convert the coordinates of the POI to be searched into second coordinates with preset precision; generating a search key value by each word in the segmentation result of the POI to be searched and each coordinate in a coordinate set, wherein the coordinate set comprises: the second coordinate and a third coordinate with a distance not exceeding the searching range, wherein the third coordinate and the second coordinate have the same precision; searching the POI list corresponding to each search key value in the inverted index, and determining the searched POI list corresponding to each search key value as a candidate POI list, wherein the candidate POI is a POI in the POI list corresponding to the search key value.

10. The apparatus of claim 9, wherein the construction module is specifically configured to: amplifying the coordinates of each POI in the electronic map database according to a preset scale to generate a first coordinate, wherein the preset scale corresponds to the preset precision;

the search module is specifically configured to: amplifying the coordinates of the POI to be searched according to a preset scale to generate a second coordinate;

11. The apparatus according to claim 9, wherein the POI list stores therein the identification numbers IDs of the POIs contained in the POI list, and the IDs in the POI list are arranged in order; the determining module is specifically configured to:

12. The apparatus of claim 11, wherein the means for determining is further configured to:

before the similarity between the POI to be searched and each candidate POI is calculated, determining a public word ratio corresponding to each candidate POI according to the number of words contained in the name of the POI to be searched, the number of words contained in the name of each candidate POI and the number of public words between the POI to be searched and each candidate POI;

13. The apparatus of claim 12, wherein the determining module is specifically configured to:

R _i ＝M _i /min(A，B _i ) (1)

wherein R is _i Representing the public word ratio corresponding to the i candidate POI, wherein i is more than or equal to 1 and less than or equal to the public word to be calculatedTotal number of candidate POI of rate; m is M _i Representing the number of common words between the POI to be searched and the ith candidate POI, A represents the number of words contained in the name of the POI to be searched, B _i The number of words contained in the name representing the i candidate POI.

14. The apparatus of claim 11, wherein the determining module is specifically configured to:

S _j ＝log(D _j /(c+M _j )) (2)

15. The apparatus according to any one of claims 9-14, wherein the apparatus further comprises:

the processing module is used for carrying out standardized processing on the names of all POIs in the electronic map database before the construction module divides the names of all POIs in the electronic map database according to words;

the processing module is further configured to perform standardized processing on the name of the POI to be searched before the searching module performs word segmentation on the name of the POI to be searched.

16. The apparatus of claim 15, wherein the normalization process comprises at least one of:

performing case, full half angle and/or complex conversion on the name;

deleting other punctuation marks without brackets in the name;

17. A POI search device, comprising: a memory and a processor, the memory for storing a computer program; the processor is configured to perform the method of any of claims 1-8 when the computer program is invoked.

18. A computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the method according to any of claims 1-8.