KR101348877B1 - Apparatus and method for set－based searching - Google Patents

Abstract

Set-based retrieval apparatus and method are disclosed. A set-based retrieval method may include maintaining a POI database storing a plurality of words, and corresponding to each syllable constituting the plurality of words, a first of the words of the POI database including the respective syllables. Maintaining an inverse array database for storing information, in response to input of a query word, identifying first information corresponding to a syllable constituting the query word with reference to the inverse array database, and common among the identified first information And outputting a word of the first information to be a search word.

Description

Set-based search device and method {APPARATUS AND METHOD FOR SET-BASED SEARCHING}

Embodiments of the present invention relate to a technique for extracting and providing a search word associated with a query by using syllables constituting a query.

Query search technology is applied in various fields (eg, web search, navigation, etc.). Conventional query retrieval techniques can extract and provide a search word including a part of an input query word from a database.

In this case, the conventional query search technique (eg, an alias technique) searches and provides all search words including a query as a part, thereby providing a search word having a relatively low search preference, and providing a search word. The search time or the search amount to extract is large.

Accordingly, there is a need for a technique for more efficiently extracting and providing a search word having a relatively high search preference.

The present invention includes a query word as a part by checking information on a word corresponding to each syllable constituting a query word in a database, and providing a word of information common among information about the checked word as a search word associated with the query word. Compared to the existing search technology that searches and provides all search words that are searched for, the purpose of the present invention is to efficiently search for and provide search words with a relatively high search preference while matching the syllables of the query word with less searching time or search volume. do.

According to the present invention, when there are a plurality of search words, the higher the degree of correspondence between the order between the syllables in the first syllable pair constituting the search word and the order between the syllables in the second syllable pair constituting the query word, the higher priority is output. In addition, the purpose of the present invention is to make it easier to recognize a search word having a syllable arranged closest to a syllable array of a query word.

In order to achieve the above object, a set-based retrieval method includes maintaining a POI database storing a plurality of words, and corresponding to each syllable constituting the plurality of words, wherein the POI database includes each syllable. Maintaining an inverse array database that stores first information about a word of, identifying a first information corresponding to a syllable constituting the query by referring to the inverse array database, in response to input of a query; The method may include outputting a word of first information that is common among the identified first information as a search word.

In another embodiment, the set-based search method may include maintaining a POI database that stores a plurality of words, and corresponding to each vowel tuple constituting the plurality of words, wherein the POI includes the respective vowel tuples. Maintaining a vowel tuple inverse array that stores words in the database;

Maintaining a vowel triple inverse array database for storing words of the POI database including the vowel triples corresponding to each vowel triple constituting the plurality of words, and in response to input of a query, the vowel tuple The method may include outputting a word corresponding to the query word as a search word by referring to an inverse array database and a vowel triple inverse array database.

The set-based retrieval apparatus implementing the method may include a POI database storing a plurality of words and a word relating to a word of the POI database including the syllables corresponding to each syllable constituting the plurality of words. A processor that identifies first information corresponding to the syllables constituting the query word by referring to the inverse array database storing the first information, and in response to input of the query word, a first common word among the identified first information; 1 may include an output unit for outputting a word of information as a search word.

In another embodiment, the set-based retrieval apparatus may further include a POI database that stores a plurality of words, and a POI database including the respective vowel tuples corresponding to each of the vowel tuples constituting the plurality of words. A vowel tuple inverse array database for storing words, a vowel triple inverse array database for storing words of the POI database including the vowel triples corresponding to each of the vowel triples constituting the plurality of words, and an input of a query word. In response, the vowel tuple inverse array database and the vowel triple inverse array database may be referred to and include an output unit for outputting a word corresponding to the query word as a search word.

According to an embodiment of the present invention, by confirming information about a word corresponding to each syllable constituting a query word in a database, and providing a word of information common among the information about the checked word as a search word associated with the query word, Compared to the existing search technology that searches and provides all search words including the query as a part, the search word or search volume is optimized, and the search words with high search preference are relatively efficiently matched with the optimal syllable of the query syllable. Can provide.

According to an embodiment of the present invention, when there are a plurality of search words, the degree of agreement between the syllables in the first syllable pair constituting the search word and the order between the syllables in the second syllable pair constituting the query word The larger the first value, the easier it is to recognize the search words arranged syllables closest to the syllable array of the query.

1 is a diagram illustrating a configuration of a set based search apparatus according to an exemplary embodiment of the present invention.
2 is a diagram illustrating an example of word word identifier and syllable identifier assignment for each word in a set based search apparatus according to an exemplary embodiment of the present invention.
3 is a diagram illustrating an example of a word identifier configuration corresponding to a syllable identifier in a set based search apparatus according to an embodiment of the present invention.
4 is a diagram illustrating an example of a search based on a query input in a set-based search apparatus according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating an example of word word and vowel identifier assignment for each word in a set based search apparatus according to an exemplary embodiment of the present invention.
6 is a diagram illustrating an example of a word identifier configuration corresponding to a vowel identifier in a set based search apparatus according to an exemplary embodiment of the present invention.
FIG. 7 illustrates another example of searching based on a query input in a set-based searching apparatus according to an embodiment of the present invention.
8 is a diagram illustrating a search word output order in a set-based search apparatus according to an embodiment of the present invention.
9 is a flowchart illustrating a set-based search method according to an embodiment of the present invention.

Hereinafter, a set-based searching apparatus and method according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a configuration of a set based search apparatus according to an exemplary embodiment of the present invention.

Referring to FIG. 1, the set-based retrieval apparatus 100 according to an embodiment of the present invention includes an allocation unit 101, a processor 103, an output unit 105, and a database 107.

First, the database 107 relates to a POI database 109 for storing a plurality of words, and a word for the word in the POI database 109 including the respective syllables, corresponding to each syllable constituting the plurality of words. It may include a reverse array database 111 for storing the first information.

The POI database 109 is a storage means for recording a name (name) of a point of interest as a word. For example, the POI database 109 may store words such as a trade name and a place name. Words of the POI database 109 are assigned unique word identifiers by the assignment unit 101 to be described later to support rapid retrieval.

The inverse array database 111 is storage means for sorting and recording first information about a word based on syllables constituting a word stored in the POI database 109. Here, the first information may be a word identifier assigned to each of the plurality of words. The syllable of the inverse array database 111 may be assigned a unique syllable identifier by the allocator 101. Therefore, in one embodiment, the inverse array database 111 stores information for identifying a word including a syllable in a form in which the word identifier is recorded corresponding to the syllable identifier.

The allocator 101 assigns a word identifier to each word stored in the POI database 109 and assigns a syllable identifier to each syllable in the inverse array database 111 that arranges and stores the syllables constituting the word. .

When a query is input, the processor 103 checks the syllable identifier for each syllable constituting the input query word and a word identifier corresponding to the syllable identifier in the database 107 in association with the query input. That is, the processor 103 may identify the first information corresponding to the syllables constituting the query by referring to the inverse array database 111 in response to the input of the query. Through this, the processor 103 extracts a word of the first information that is common among the identified first information (ie, a word identifier identified using a syllable identifier) from the POI database 109 and outputs it as a search word. It is possible to provide an environment in which the output unit 105 is controlled.

The processor 103 provides an environment in which a search word can be derived by using a word identifier associated with a syllable, so that even when a query word in abbreviated form is input, rather than the full name of a business name to be searched, the processor name to be searched is input. By providing a search word associated with, the query input is made available.

For example, the processor 103 may be a form in which "Kim Jung-yeol Classical Guitar Studio" is abbreviated, and "Kim Jung-yeol Classical Guitar Studio" may be provided as a search word when a query of "Kim Kyeol-gi Classical Studio" is input.

In the identification of the first information, the processor 103 may identify only the first information corresponding to the syllable “all” constituting the query word. That is, the processor 103 may retrieve the first information corresponding to all syllables of the query word from the inverse array database 111.

As another example of identifying the first information, the processor 103 may identify first information corresponding to at least some of the syllables constituting the query word. In other words, the processor 103 can retrieve, for example, the first information of the word registered as the abbreviation from the inverse array database 111.

The output unit 105 outputs, as a search word, a word of a common word identifier among the word identifiers identified by the syllable identifier by the processor 103. That is, the output unit 105 outputs a word of the first information that is common among the identified first information as a search word.

Under the condition that the word identifiers for each of the plurality of words are stored as the first information in the inverse array database 111, the processor 103 may select word identifiers stored in correspondence with syllables constituting the query word. Identifying at 111, the output unit 105 may output, from the POI database 109, a word designated by a word identifier common among the identified word identifiers as a search word.

According to another embodiment of the present invention, the output unit 105 according to the output number of the plurality of search words, according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word The search word given a weight and a relatively high weight may be given priority. That is, the output unit 105 may output the user's gaze on the screen with a greater number of matches between the first syllable constituting the search word and the second syllable constituting the query word.

Alternatively, the output unit 105 assigns weights in consideration of the matching number of the combination of the first syllable and the combination of the second syllable, and preferentially outputs the search word given the relatively high weight. Can be. That is, the output unit 105 is a screen where the gaze of the user becomes greater as the degree of agreement between the syllables in the first syllable pair constituting the search word and the order between the syllables in the second syllable pair constituting the query word increases. Can be output to the

On the other hand, even when implemented in the embodiment described above, if the output of the search word is not possible (that is, there is no common first information), the set-based search apparatus 100 of the present invention by using a combination of query words You can output the search word.

To this end, the set-based search apparatus 100 may reconstruct the database 107. That is, the retrieval apparatus 100 reconstructs the database 107 including the vowel tuple inverse array database 113 for two vowel combinations and the vowel triple inverse array database 115 for three vowel combinations.

The vowel tuple inverse array database 113 stores second information about words in the POI database 109 including the respective vowel tuples, corresponding to each vowel tuple constituting a plurality of words. Here, the vowel tuple refers to a vowel (neutral) combination of two consecutive syllables in a word. For example, the vowel tuple may be 'TT ㅖ', 'ㅖ ㅡ' and the like for the word 'relay star class'.

The vowel triple inverse array database 115 stores second information about words in the POI database 109 including the vowel triples corresponding to each vowel triple constituting a plurality of words. Here, the vowel triple refers to a vowel (neutral) combination of three consecutive syllables in a word. For example, the vowel triple may be 'TT ㅖ ㅡ' or 'ㅖ ㅡ ㅏ' for the word 'relay star class'.

When the database 107 is reconfigured, the allocator 101 allocates a vowel identifier to each of the vowel tuples and the vowel triples and stores them in the vowel tuple inverse array database 113 and the vowel triple inverse array database 115.

At this time, if there is another vowel in the vowel combination (vowel tuple and vowel triple) included in the vowel combination (vowel tuple and vowel triple), the assigning unit 101 may change the vowel identifier with the vowel combination. That is, the allocator 101 may set a plurality of predetermined vowels as one representative vowel in the vowel tuple inverse array database 113, the vowel triple inverse array database 111, and the query word. For example, the allocator 101 may assign a single vowel identifier by treating vowel tuples 'ㅏ' and 'ㅏ ㅏ' equally.

After reconstructing the database 107, the processor 103 may identify a vowel identifier for each vowel combination of syllables constituting the query word and a word identifier corresponding to the vowel identifier in the database 107. That is, the processor 103 refers to the vowel tuple inverse array database 113 and the vowel triple inverse array database 115 to identify second information corresponding to the syllables constituting the query word. Here, the second information may be a word identifier assigned to each of the plurality of words in the same manner as the first information described above.

In the identification of the second information, the processor 103 matches a vowel tuple or vowel triple corresponding to each of the left vowel combinations associated with up to m syllables (the m is a natural number) based on the starting syllable of the query word. Identifying the second information corresponding to the vowel tuple inverse array database 113 or the vowel triple inverse array database 115, and correspondingly, n times (n is a natural number) based on an end syllable of the query word. Second information corresponding to the vowel triple or the vowel tuple corresponding to each of the right vowel combinations associated with the syllables of the second syllable may be identified from the vowel tuple inverse array database 113 or the vowel triple inverse array database 115.

The output unit 105 is a set-based search apparatus for outputting a word of the second information common in the identified second information as a search word.

More specifically, the processor 103 refers to the vowel tuple inverse array database 113 and matches a vowel matching left vowel combination associated with the second syllable (m = 2) based on the starting syllable of the query word. Identify the first candidate word corresponding to the tuple and match the right vowel combination that is associated with the third (n = 3) syllable based on the end syllable of the query by referring to the vowel triple inverse array database 115. A second candidate word corresponding to the vowel triple is identified, and a word common to the first candidate word and the second candidate word is identified as a third candidate word.

Similarly, the processor 103 refers to the vowel triple inverse array database 115 to a vowel triple that matches the left vowel combination that is associated with the third syllable (m = 3) based on the beginning syllable of the query. A vowel that identifies a corresponding fourth candidate and matches the right vowel combination that is associated with the second (n = 2) syllable based on the end syllable of the query by referring to the vowel tuple inverse array database 113 A fifth candidate word corresponding to the tuple is identified, and a word common to the fourth candidate word and the fifth candidate word is identified as the sixth candidate word. In this case, the third candidate word and the sixth candidate word are output through the output unit 105 as a search word.

Under the condition that the word identifiers for each of the plurality of words are stored as the second information in the vowel tuple inverse array database 113 and the vowel triple inverse array database 115, the processor 103 is each in syllables constituting the query word. Correspondingly stored word identifiers are identified in the vowel tuple inverse array database 113 and the vowel triple inverse array database 115, and the output unit 105 identifies a word designated by a common word identifier among the identified word identifiers. It can be output as a search word from the POI database 109.

On the other hand, if there is a plurality of search words outputted, the output unit 105 increases the matching score by L according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word. And the number of syllables that match the initial and neutral, or the neutral and final, of the first syllable and the second syllable, except for the syllable associated with the increase of the matching score in the first stage. Accordingly, a second step of increasing the matching score by M, and a step of decreasing the matching score by N according to the number of the first syllables remaining except for the syllable associated with the increase of the matching score in the first and second stages. Step 3 is performed, and the search words with a relatively high sum of matching scores are given priority. At this time, L is larger than M, and M is set larger than N.

In other words, when there are a plurality of words (search word) of the common word identifier among the word identifiers identified using the vowel identifier, the output unit 105 matches the second word that constitutes the query word. After the matching score is increased (+30 points) for the first syllable, the matching score is increased (+15 points) for the first syllable having the same neutrality as that of the second syllable and having the same initial or finality. Can be given. In this case, the output unit 105 subtracts (-3 points) the matching score from the first syllable that is different from the second syllable or the neutral, or the same syllable as the second syllable but the first or the last is different. In addition, the search word having a relatively high matching score may be prioritized and output.

In the following, a unique identifier (word identifier, syllable identifier, vowel identifier) is assigned to each word, syllable, or vowel combination stored in each DB in the database 107 to illustrate that the invention is implemented. It is obvious that the embodiments are only for convenience and not for limiting the embodiments. For example, the invention may be implemented such that a word, syllable, or vowel combination itself is extracted directly from each DB without additional identifier assignment, which is within the scope of the present invention.

2 is a diagram illustrating an example of word word identifier and syllable identifier assignment for each word in a set based search apparatus according to an exemplary embodiment of the present invention.

Referring to FIG. 2, the set-based search apparatus 100 assigns a word identifier to each word stored in the POI database 109. For example, the set-based retrieval apparatus 100 may have a word stored in the POI database 109 such as "relay star class, Kim Jung-yeol classical guitar studio, life time fitness club,... , 'Gumgangsan Ribs' is stored, 'I' is a word identifier of 'A' in 'Relay Star Class', 'B' is identified in 'Kim Jung-yeol Classical Guitar Studio', 'C' is identified in 'Lifetime Fitness Club', The word identifier of 'K' can be assigned to 'Kumgangsan Ribs'.

Subsequently, the set-based retrieval apparatus 100 allocates a syllable identifier to each syllable constituting a word and stores the syllable identifier in the inverse array database 111. For example, the set-based retrieval apparatus 100 may each of the syllables constituting the 'relay star class', that is, 'middle', 'system', 's', 'ta', 'cle', 'ra', 's' For, we can assign syllable identifiers.

In the assignment of syllable identifiers, the set-based search apparatus 100 sorts each syllable constituting each word based on a set order (for example, vowel and consonant order), and distinguishes syllable identifiers according to the sorted order. Can be assigned. For example, the set-based retrieval apparatus 100 may include each word stored in the POI database 109, namely, 'relay star class', 'Kim Jung-yeol classical guitar studio', 'life time fitness club',... , Each syllable constituting 'Kumgangsan Ribs' is sorted based on vowel and consonant order, the syllable identifier of '0' in the first sorted 'gal', and the syllable identifier of '1' in the 'gang' arranged second You can assign a syllable identifier of '2' to the third sorted 'gun'.

3 is a diagram illustrating an example of a word identifier configuration corresponding to a syllable identifier in a set based search apparatus according to an embodiment of the present invention.

Referring to FIG. 3, the set-based retrieval apparatus 100 maintains a syllable identifier associated with a syllable and a syllable identifier associated with a word in the inverse array database 111 based on the syllable identifier of a syllable in a query word that is input later. This allows the identification of word identifiers.

In detail, the set-based retrieval apparatus 100 may store the word identifier for the word including the syllable matching the syllable identifier with respect to the syllable identifier. For example, the set-based retrieval apparatus 100 may include a syllable identifier of 'O', a word identifier of a word of 'Keumgangsan rib' including a syllable matching 'O', that is, 'gal', that is, 'K'. The reverse array database 111 can be configured to store correspondingly. In addition, the set-based retrieval apparatus 100 includes a syllable identifier of '1', a syllable matching '1', that is, a word of 'vitamin health club' and 'geumgang ribs' including 'river'. The reverse identifier database 111 may be configured such that the word identifiers 'E' and 'K' are stored correspondingly.

4 is a diagram illustrating an example of a search based on a query input in a set-based search apparatus according to an embodiment of the present invention.

Referring to FIG. 4, the set-based retrieval apparatus 100 interlocks a syllable identifier for each syllable constituting the input query word and a word identifier corresponding to the syllable identifier in the inverse array database 111 in association with a query input. After confirming, the words of the word identifiers common among the identified word identifiers can be output from the POI database 109 as search words.

For example, the set-based search apparatus 100 may include a syllable identifier for 's', that is, a word identifier of 'A, B, C, D, I' corresponding to '23' when a query of 'star class' is input. ('Set 1') is checked in the inverse array database 111, and a syllable identifier for 'other', that is, a word of 'A, B, C, D, E, F, H, I' corresponding to '44' The identifier ('set 2') can be found in the inverse array database 111.

The set-based retrieval apparatus 100 checks word identifiers associated with each of 'cle', 'rae', and 's', and identifies the common word identifiers of the identified word identifiers, that is, words 'A and B'. The relay star class, B: Kim Jung-yeol Classical Guitar Studio 'can be identified and output from the POI database 109 as a search word.

FIG. 5 is a diagram illustrating an example of word word and vowel identifier assignment for each word in a set based search apparatus according to an exemplary embodiment of the present invention.

Referring to FIG. 5, the set-based search apparatus 100 assigns a word identifier to each word stored in the POI database 109. For example, the set-based retrieval apparatus may store the word 'relay star class, cluetime square, sparkle center, vitamin health club, and Geumgang ribs' stored in the POI database 109 in the 'relay star class'. Word identifier of 'A', word identifier of 'B' at 'Clue Time Square', word identifier of 'C' at 'Sparkle Car Center', word identifier of 'D' at 'Vitamin Health Club', 'Kumgang Ribs' You can assign a word identifier of 'E'.

Thereafter, the set-based retrieval apparatus 100 may assign a vowel identifier to each vowel combination of syllables constituting a word. In this case, the set-based retrieval apparatus 100 may allocate a vowel identifier for each vowel combination of a plurality of adjacent syllables.

For example, the set-based retrieval apparatus 100 may allocate a vowel identifier for each of the two syllable vowel combinations constituting the relay star class and store the vowel identifier in the vowel tuple inverse array database 113. That is, the set-based retrieval apparatus 100 assigns a syllable identifier to a vowel combination of 'relay', that is, 'grace', and assigns a vowel combination of 'ges', that is, 'yes' and 'star'. Ie the vowel combination of 'a' and 'tackle', that is, the vowel combination of 'a' and 'class', ie the vowel combination of 'a' and 'class', i.e. Can be assigned.

In addition, the set-based retrieval apparatus 100 may allocate a vowel identifier to each of the three syllable vowel combinations constituting the relay star class and store the vowel identifier in the vowel triple inverse array database 115. That is, the set-based retrieval apparatus 100 allocates a syllable identifier to a vowel combination of 'relays', that is, 'yeye', and assigns a vowel combination of 'gaestar', that is, 'yes', 'star'. You can assign a vowel identifier for each of the vowel combinations of the 'cle', ie 'a', 'the vowels' of the tackle' have.

When assigning a vowel identifier, the set-based search apparatus 100 may assign a vowel identifier by changing to a vowel combination having the other vowel when there is another vowel in the vowel combination included in the vowel combination. That is, the set-based retrieval apparatus 100 may process a plurality of predetermined vowels as one representative vowel, and, for example, make a vowel having the same pronunciation as one representative vowel to facilitate input and search. do.

For example, the set-based retrieval apparatus 100 sets the vowel combination of `` ae '' to vowels of `` ae '' when there are other vowels of `` e '' and hom included in the vowel combination of `` ae ''. You can change to a combination and assign a vowel identifier to the changed vowel combination of 'Ae'.

Subsequently, the set-based retrieval apparatus 100 may sort each vowel combination of syllables constituting each word based on a set order (for example, vowel order), and allocate a distinct vowel identifier according to the sorted order. have.

The set-based retrieval apparatus 100 is, for example, each word stored in the POI database 109, that is, 'relay star class', 'clue time square', 'sparkle car center', 'vitamin health club', 'geumgangsan rib' For example, each of the two syllable vowel combinations is arranged based on the vowel order, and the vowel identifier of '0' in the first sorted 'ah' and the vowel identifier of '1' in the second sorted '아아' By assigning the vowel tuple inverse array database 113 can be configured.

In addition, the set-based retrieval apparatus 100 sorts the vowel combination for each of the three syllable vowel combinations for each word, similarly to the vowel combination of the two syllables, and sets '0' to the first sorted 'aa'. A collection triple inverse array database 115 may be configured by assigning a collection identifier of and assigning a collection identifier of '1' to a second sorted 'eye'.

6 is a diagram illustrating an example of a word identifier configuration corresponding to a vowel identifier in a set based search apparatus according to an exemplary embodiment of the present invention.

Referring to FIG. 6, the set-based retrieval apparatus 100 correlates and stores a vowel identifier associated with a word of the POI database 109 and a vowel identifier of a vowel combination in association with each other. The database 107 can be configured to include an inverse array database 115.

In detail, the set-based search apparatus 100 may store a word identifier for a word including a vowel combination matching the vowel identifier with respect to the vowel identifier.

For example, the set-based retrieval apparatus 100 includes a vowel combination matching 'O', that is, 'A', for a vowel identifier of 'O' in a vowel identifier corresponding to two syllable vowel combinations. The vowel tuple inverse array database 113 may be configured such that a word identifier for a word of 'Kumgangsan ribs', that is, 'E' is correspondingly stored. In addition, the set-based search apparatus 100 includes a vowel combination matching '1', that is, a word of 'sparkle car center' including 'a' (or 'a') for a vowel identifier of '1'. The vowel tuple inverse array database 115 may be configured such that a word identifier for the word 'C' is correspondingly stored.

Similarly, the set-based search apparatus 100 includes a vowel combination matching 'O', that is, 'A', for the vowel identifier of 'O' in a vowel identifier corresponding to three syllable vowel combinations. The vowel triple inverse array database 115 may be configured such that a word identifier, ie, 'E', of the word of 'Kumgangsan Ribs' is correspondingly stored. In addition, the set-based search apparatus corresponds to a vowel identifier of '1', a word identifier of 'void combination of' 1 ', that is, a word identifier of a word of' Kumgangsan rib 'including' Ai '. The collection triple inverse array database 115 may be configured to be stored and stored.

FIG. 7 illustrates another example of searching based on a query input in a set-based searching apparatus according to an embodiment of the present invention.

Referring to FIG. 7, the set-based retrieval apparatus 100, in association with a query input, includes a syllable identifier for each syllable constituting the input query word and a word identifier corresponding to the syllable identifier. And the vowel triple inverse array database 115, and if there is no common word identifier among the identified word identifiers, the search word may be output using the vowel identifier for each vowel combination of syllables constituting the query word. .

In this case, the set-based retrieval apparatus 100 is a vowel tuple of word identifiers corresponding to vowel identifiers for each of the left vowel combinations that are associated with the syllable up to m (the m is a natural number) based on the starting syllable of the query. The vowel identifier for each of the right vowel combinations identified in the inverse array database 113 or the vowel triple inverse array database 115 or associated with the syllable up to n (where n is a natural number) based on the ending syllable of the query word. The corresponding word identifier may be checked in the vowel tuple inverse array database 113 or the vowel triple inverse array database 115.

Referring to FIGS. 6 and 7, for example, the set-based search apparatus 100 may input a syllable up to the second syllable, that is, “ka-kup” and end based on the start syllable, when a query word of “ka-kappa spak” is input. Based on the syllables, the word identifiers for the third syllable, that is, each of the 'sparks' may be checked in each of the vowel tuple inverse array database 113 and the vowel triple inverse array database 115.

In other words, the set-based retrieval apparatus 100 changes the left vowel combination of the 'Catetta Spark', that is, the vowel combination 'Ae' of the 'Cat' to a vowel combination 'Ae' with the vowels of the homophone, and changes the ' The vowel identifier for Aeah, that is, the word identifier ('set 1') of 'C' corresponding to '1' may be checked in the vowel tuple inverse array database 113.

Subsequently, the set-based retrieval apparatus 100 is a right side vowel combination of 'Cacent's Park', that is, a vowel identifier for 'Ea', that is, a word identifier ('set) of' A, C 'corresponding to' 15 '. 2 ′) can be found in the vowel triple inverse array database 115.

The set-based retrieval apparatus 100 identifies the common word identifier 'C' of sets 1 and 2 and identifies the word 'sparkle center' corresponding to the word identifier 'C' as the POI database 107. Can be output as a search word.

Also, the set-based retrieval apparatus 100 may have a third syllable based on the start syllable, that is, the second syllable based on the 'cañata' and the ending syllable, with respect to the input query word of 'Chapteraspark'. The word identifiers for the syllables, ie, 'tags', may be further identified from the vowel tuple inverse array database 113 and the vowel triple inverse array database 115, respectively.

That is, the set-based retrieval apparatus 100 changes the left vowel combination of the 'capitaspark', that is, the vowel combination 'aea' of the 'cacenta' into a vowel combination 'aae' with the same vowel, A vowel identifier for 'Ah Aya', that is, a word identifier of 'C' corresponding to '2' ('set 3') may be checked in the vowel triple inverse array database 115.

Subsequently, the set-based retrieval apparatus 100 is a right-sided combination of 'Casenta Stark', that is, a vowel identifier for 'A', that is, a word identifier ('set) of' A, C, D 'corresponding to' 2 '. 4 ′) can be found in the vowel tuple inverse array database 113.

The set-based search apparatus 100 checks the common word identifier 'C' of sets 3 and 4, identifies the word 'sparkle center' corresponding to the word identifier 'C' as the POI database 107, and Can be output as a search word.

In the above example, the word identifier common among 'set 1' and 'set 2' and the common word identifier among 'set 3' and 'set 4' are the same, so that one search word 'sparkle center' is outputted. However, when common word identifiers are different from each other, a plurality of search words may be identified and output.

8 is a diagram illustrating a search word output order in a set-based search apparatus according to an embodiment of the present invention.

Referring to FIG. 8, when the set-based search apparatus 100 is linked to a query input and outputs a plurality of search words, i) weights associated with syllables or ii) matching scores associated with initial / neutral / final in syllables And outputting a search word given a relatively high weight or matching score (ie, in descending order), it is easier to recognize a search word having a high search preference.

Iii) In case of outputting based on the weight associated with syllable,

The set-based search apparatus 100 assigns weights according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word, and prioritizes the search word given a relatively high weight. You can print In this case, the set-based retrieval apparatus 100 may give the weight in consideration of the number of matches between the combination of the first syllable and the combination of the second syllable.

For example, the set-based search apparatus 100 includes a syllable constituting each search word and syllables constituting each search word when the query word is “other”, and the output search words are “Kim Jung-yeol classical guitar” and “percussion instrument lecture”. As two syllables, i.e., the other and the same, are equal, weights may be equally assigned to each search word. In this case, the set-based retrieval apparatus 100 includes a 'guitar' in which a syllable of syllables is combined is included in the 'kim jeongyeol classical guitar', while not included in the 'percussion instrument lecture,' By adding a weight to the search word of 'percussion lecture', the search word of 'Kim Jungyeol classical guitar' can be given a relatively higher weight.

Accordingly, the set-based retrieval apparatus 100 may output the Kim Jong-yeol classical guitar and the percussion lecture order based on the weight.

Ii) In case of outputting based on the matching score associated with the initial / neutral / finality in the syllable,

The set-based retrieval apparatus 100 may match the second syllable constituting the search word with a matching score for the first syllable that matches the second syllable constituting the query word in response to the input of the query word. After increasing, the matching score may be increased by increasing the matching score with respect to the first syllable having the same neutrality as that of the second syllable and having the same initial or finality. Subsequently, the set-based retrieval apparatus 100 may subtract a matching score from a first syllable different from the second syllable or the first syllable having the same neutrality as the second syllable but different from the initial or the finality.

For example, the set-based search apparatus 100 may have a first syllable and a 'sparkle casenta' when the query is 'cachetta park' and gives a matching score for 'sparkle car center' among a plurality of search words. Since four syllables of the second syllables constituting the Cajuntas Spark, that is, the syllables of 'S, Fah, Ka, Ta', are matched, when one syllable is matched, a score of '30' is given. '120' may be given as a matching score for the 'centre'.

Subsequently, the set-based retrieval apparatus 100 uses matching syllables except for the four syllables in each syllable constituting the query word of 'Capteta Spark' and syllables constituting the search word of 'Sparkle Car Center'. You can increase or decrease it. That is, the set-based search apparatus 100 may have the same neutrality ('e') and the same finality ('b') based on the 'sy' syllable constituting the search word based on the 'sy' syllable constituting the query word. In this case, a matching score of '135' may be given by increasing '15'. In addition, the set-based search apparatus 100 may have the same neutral (')' and the same initial ('ㅋ') based on the 'large' syllable that constitutes the query word. In this case, '15' may be increased to give a matching score of '150'.

In addition, the set-based search apparatus 100 has a first syllable and a 'car' that constitute a 'sparkle car center' when the query word is 'capita spark' and a matching score for 'spaghetti' is among the plurality of search words. Since two syllables of the second syllable constituting the pentaspark, that is, the syllable 'spa' syllables match, a matching score of '30' is given when one syllable matches, thus matching the 'sparkle car center' As a score, '60' can be given.

Subsequently, the set-based search apparatus 100 has the same neutral ('e'), but different initial and / or final, based on '하는' syllables constituting the search word. By subtracting 3 ', a matching score of' 57 'can be given.

In addition, the set-based search apparatus 100 has a '3' because the 'T' syllable constituting the search word is different from the neutral ('yi') based on the 'ka, ta, k' syllable constituting the query word. By subtracting, a matching score of '54' can be given.

The set-based search apparatus 100 first outputs a search word given a relatively high weight or matching score among a plurality of search words (ie, a descending order), so that the search word having a high search preference is easily recognized. To do it.

Accordingly, the set-based search apparatus 100 may output the sparkle center and the spaghetti based on the matching score.

9 is a flowchart illustrating a set-based search method according to an embodiment of the present invention.

Referring to FIG. 9, in step 901, the set-based search apparatus 100 assigns a word identifier to each word stored in the POI database 109, and assigns a syllable identifier to each syllable constituting the word. .

In addition, the set-based retrieval apparatus 100 may assign a vowel identifier to each vowel combination of syllables constituting a word. In this case, the set-based retrieval apparatus may allocate a vowel identifier to each of the adjacent vowel combinations of a plurality of syllables, and if there are different vowels with the same vowel included in the vowel combination, change the vowel combination with the different vowels. You can assign a vowel identifier.

In step 903, the set-based search apparatus configures the database 107 so that the word identifiers are stored in correspondence with the syllable identifiers. The set-based retrieval apparatus 100 may configure the inverse array database 111 so that a word identifier for a word including a syllable matching the syllable identifier corresponds to the syllable identifier.

In another embodiment, the set-based retrieval apparatus 100 may configure the vowel tuple inverse array database 113 and the vowel triple inverse array database 115 by assigning a vowel identifier to each vowel combination of syllables constituting a word. .

In operation 905, the set-based retrieval apparatus 100 checks the syllable identifier and the word identifier corresponding to the syllable identifier in the inverse array database 111 in association with the query input.

In operation 907, the set-based search apparatus 100 searches for a word of a word identifier that is common among the identified word identifiers in the POI database 109 and outputs the search word.

The set-based search apparatus 100 assigns weights according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word when the output search word is plural, and the weight A search word given a relatively high value may be given priority. In this case, the set-based retrieval apparatus 100 may give the weight in consideration of the number of matches between the combination of the first syllable and the combination of the second syllable.

In step 907, when there is no common word identifier, the set-based search apparatus 100 performs a vowel identifier for each vowel combination of syllables constituting a query word, and a word identifier corresponding to the vowel identifier. In the database 113 and the vowel triple inverse array database 115, a word of a common word identifier among the identified word identifiers may be output as a search word.

In this case, the set-based retrieval apparatus 100 is a word identifier corresponding to a vowel identifier for each of the left vowel combinations that are associated with the syllable up to m (the m is a natural number) based on the starting syllable of the query word. In the database, a word identifier corresponding to a vowel identifier for each right vowel combination that is associated with a syllable up to n times (n is a natural number) based on the ending syllable of the query word may be checked in the database. .

In addition, the set-based retrieval apparatus 100 may identify a vowel included in the vowel combination and a vowel identifier assigned to the vowel combination having the same vowel.

The set-based search apparatus 100 matches the search word that matches a second syllable constituting the query word when there are a plurality of words of a word identifier that is common among the word identifiers identified using the vowel identifier, that is, the search words. The matching score may be increased for the first syllable, and then the matching score may be increased for the first syllable having the same neutrality as that of the second syllable and having the same initial or finality. In this case, the set-based retrieval apparatus subtracts the matching score from the first syllable different from the second syllable or the first syllable having the same neutrality as that of the second syllable but differs in the initial or finality, and provides the matching score. Relatively high search words can be given priority.

Embodiments of the present invention may be implemented in the form of program instructions that can be executed on various computer means and recorded on a computer readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above embodiments, and those skilled in the art to which the present invention pertains various modifications and variations from such descriptions. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the equivalents of the claims, as well as the claims.

100: set-based search device
101: allocation unit
103: processor
105: output unit
107: database
109: POI Database
111: inverse array database
113: vowel tuple inverse array database
115: vowel triple inverse array database

Claims (34)

Maintaining a POI database storing a plurality of words;
Maintaining an inverse array database storing first information about a word of the POI database including each syllable, corresponding to each syllable constituting the plurality of words;
In response to input of a query word, referring to the inverse array database, identifying first information corresponding to a syllable constituting the query word; And
Outputting a word of the first information including the syllables constituting the query word among the identified first information as a search word;
Set-based search method comprising a. The method of claim 1,
Identifying the first information,
In response to the input of the query, identifying only the first information corresponding to the syllable “all” constituting the query by referring to the inverse array database.
Set-based search method comprising a. The method of claim 1,
Identifying the first information,
In response to the input of the query, identifying first information corresponding to at least some of the syllables constituting the query by referring to the inverse array database.
Set-based search method comprising a. The method of claim 1,
When the output search word is plural,
Outputting the word of the first information as a search word,
Outputting the first syllable constituting the search word and the second syllable constituting the query word as the number of matches is higher
Set-based search method comprising a. The method of claim 1,
When the output search word is plural,
Outputting the word of the first information as a search word,
Outputting a larger order of matching between the syllables within two adjacent syllables constituting the search word and the order between the syllables within two adjacent syllables constituting the query word;
Set-based search method comprising a. The method of claim 1,
The inverse array database stores word identifiers for each of the plurality of words as the first information.
Identifying the first information,
Identifying word identifiers stored in correspondence with syllables forming the query word in the inverse array database;
Lt; / RTI >
Outputting the word of the first information as a search word,
Outputting, from among the identified word identifiers, a word designated by a word identifier common to the word identifier of each syllable constituting the query, from the POI database as a search word;
Set-based search method comprising a. The method of claim 1,
In case there is no output search word,
Maintaining a vowel tuple inverse array database that stores second information about a word in the POI database including the vowel tuples constituting the plurality of words;
Maintaining a vowel triple inverse array database that stores second information about words in the POI database including the vowel triples corresponding to each vowel triple constituting the plurality of words;
Identifying second information corresponding to a syllable constituting the query by referring to the vowel tuple inverse array database and the vowel triple inverse array database; And
Outputting a word of second information in common among the identified second information as a search word;
Set-based search method further comprising. The method of claim 7, wherein
Identifying the second information,
A second information corresponding to a vowel tuple or a vowel triple corresponding to each of the left vowel combinations associated with syllables up to m (the m is a natural number) based on the starting syllable of the query word; Identifying from the vowel triple inverse array database; And
The second information corresponding to the vowel triple or the vowel tuple corresponding to each of the right vowel combinations associated with the syllable up to n times (n is a natural number) based on the ending syllable of the query word, the vowel triple inverse array database or Identifying from the vowel tuple inverse array database
Set-based search method comprising a. The method of claim 7, wherein
The vowel tuple inverse array database and the vowel triple inverse array database store word identifiers for each of the plurality of words as the second information.
Identifying the second information,
Identifying word identifiers stored in correspondence with syllables constituting the query in the vowel tuple inverse array database and the vowel triple inverse array database
Lt; / RTI >
Outputting the word of the second information as a search word,
Outputting a word designated by a common word identifier among the identified word identifiers as a search word from the POI database;
Set-based search method comprising a. The method of claim 7, wherein
When the output search word is plural,
Outputting the word of the second information as a search word,
A first step of increasing the matching score by L according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word;
The matching score is M according to the number of syllables in which the initial and the neutral syllables or the neutral and the final syllables are matched among the first syllable and the second syllable, except for the syllable associated with the increase of the matching score in the first stage. A second step of incrementing;
A third step of decreasing the matching score by N according to the number of the first syllables remaining, except for the syllables associated with the increase of the matching score in the first and second steps; And
Outputting a search word having a relatively high matching score in priority order
Set-based search method comprising a. The method of claim 10,
Wherein L is greater than M and M is greater than N. 2. Maintaining a POI database storing a plurality of words;
Maintaining a vowel tuple inverse array database for storing the words of the POI database including the vowel tuples corresponding to the vowel tuples constituting the plurality of words;
Maintaining a vowel triple inverse array database for storing the words of the POI database including the vowel triples corresponding to the vowel triples constituting the plurality of words; And
In response to an input of a query word, referring to the vowel tuple inverse array database and the vowel triple inverse array database, outputting a word corresponding to the query word as a search word;
Set-based search method comprising a. The method of claim 12,
In the vowel tuple inverse array database, the vowel triple inverse array database, and the query word, a plurality of vowels having the same predetermined Hangul pronunciation are treated as one representative vowel of the plurality of vowels. The method of claim 12,
The step of outputting a word corresponding to the query word as a search word,
Identifying a first candidate word corresponding to a vowel tuple that matches a left vowel combination associated with a second syllable based on a start syllable of the query by referring to the vowel tuple inverse array database;
Identifying a second candidate word corresponding to a vowel triple matching a right vowel combination that is associated with a third syllable based on the end syllable of the query by referring to the vowel triple inverse array database;
Identifying a word that is common to the first candidate word and the second candidate word as a third candidate word;
Identifying a fourth candidate word corresponding to a vowel triple matching a left vowel combination that is associated with a third syllable based on a starting syllable of the query by referring to the vowel triple inverse array database;
Identifying a fifth candidate word corresponding to a vowel tuple that matches the right vowel combination associated with a second syllable based on the end syllable of the query by referring to the vowel tuple inverse array database;
Identifying a word common to the fourth candidate word and the fifth candidate word as a sixth candidate word; And
Outputting the third candidate word and the sixth candidate word as the search word;
Set-based search method comprising a. The method of claim 12,
When the output search word is plural,
The step of outputting a word corresponding to the query word as a search word,
Outputting the first syllable constituting the search word and the second syllable constituting the query word in a higher order
Set-based search method comprising a. The method of claim 12,
When the output search word is plural,
The step of outputting a word corresponding to the query word as a search word,
Outputting the first syllable constituting the search word and the second syllable constituting the query word as the number of syllables that match the initial and neutral or the neutral and the final match increases
Set-based search method comprising a. The method of claim 12,
When the output search word is plural,
The step of outputting a word corresponding to the query word as a search word,
A first step of increasing the matching score by L according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word;
The matching score is M according to the number of syllables in which the initial and the neutral syllables or the neutral and the final syllables are matched among the first syllable and the second syllable, except for the syllable associated with the increase of the matching score in the first stage. A second step of incrementing;
A third step of decreasing the matching score by N according to the number of the first syllables remaining, except for the syllables associated with the increase of the matching score in the first and second steps; And
Outputting a search word having a relatively high matching score in priority order
Set-based search method comprising a. 18. The method of claim 17,
Wherein L is greater than M and M is greater than N. 2. A POI database for storing a plurality of words;
An inverse array database that stores first information about a word of the POI database including each syllable, corresponding to each syllable constituting the plurality of words;
A processor that, in response to an input of a query word, refers to the inverse array database and identifies first information corresponding to a syllable constituting the query word; And
An output unit for outputting a word of common first information including the syllables constituting the query word among the identified first information as a search word;
Set-based search device comprising a. 20. The method of claim 19,
The processor comprising:
And a set-based retrieval device for identifying only first information corresponding to the syllable "all" constituting the query. 20. The method of claim 19,
When the output search word is plural,
The output unit includes:
And a set-based retrieval device that outputs a higher order of matching between the syllables within two adjacent syllables constituting the search word and the order between the syllables within two adjacent syllables. 20. The method of claim 19,
The inverse array database stores word identifiers for each of the plurality of words as the first information.
The processor comprising:
Identify the word identifiers stored in the syllables constituting the query word in the inverse array database;
The output unit includes:
And a word set by the word identifier common to the word identifier of each syllable constituting the query word from the identified word identifiers, from the POI database as a search word. 20. The method of claim 19,
A vowel tuple inverse array database that stores second information about words in the POI database including the vowel tuples constituting the plurality of words, wherein the vowel tuples include the vowel tuples; And
A vowel triple inverse array database for storing second information about words in the POI database including the vowel triples corresponding to the vowel triples constituting the plurality of words.
Further comprising:
If there is no common word above,
The processor comprising:
Identifying second information corresponding to a syllable constituting the query by referring to the vowel tuple inverse array database and the vowel triple inverse array database,
The output unit includes:
And a set-based retrieval device for outputting a word of second information that is common among the identified second information as a search word. 24. The method of claim 23,
The processor comprising:
A second information corresponding to a vowel tuple or a vowel triple corresponding to each of the left vowel combinations associated with syllables up to m (the m is a natural number) based on the starting syllable of the query word; Identify from the vowel triple inverse array database,
The second information corresponding to the vowel triple or the vowel tuple corresponding to each of the right vowel combinations associated with the syllable up to n times (n is a natural number) based on the ending syllable of the query word, the vowel triple inverse array database or And a set based retrieval device for identifying from the vowel tuple inverse array database. 24. The method of claim 23,
The vowel tuple inverse array database and the vowel triple inverse array database store word identifiers for each of the plurality of words as the second information.
The processor comprising:
Word identifiers stored corresponding to the syllables constituting the query word are identified in the vowel tuple inverse array database and the vowel triple inverse array database,
The output unit includes:
And a word set specified by a common word identifier among the identified word identifiers, from the POI database as a search word. 24. The method of claim 23,
When the output search word is plural,
The output unit includes:
A first step of increasing the matching score by L according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word, and the syllable associated with the increase of the matching score in the first step. A second step of increasing the matching score by M according to the number of syllables in which the first and the second syllables match the initial and the neutral or the neutral and the finality of the first syllable and the second syllable, except for the first syllable; In step 2, a third step of decreasing the matching score by N according to the number of the first syllables remaining except for the syllables related to the increase of the matching score is performed, and prioritizing the search word having a relatively high matching score. Set-based search device to output. The method of claim 26,
And L is greater than M and M is greater than N. 2. A POI database for storing a plurality of words;
A vowel tuple inverse array database for storing words of the POI database including the vowel tuples constituting the plurality of words;
A vowel triple inverse array database that stores words of the POI database including the vowel triples corresponding to the vowel triples constituting the plurality of words; And
An output unit for outputting a word corresponding to the query word as a search word by referring to the vowel tuple inverse array database and the vowel triple inverse array database in response to the input of the query word
Set-based search device comprising a. 29. The method of claim 28,
In the vowel tuple inverse array database, the vowel triple inverse array database, and the query word, a plurality of vowels having the same predetermined Hangul pronunciation are processed as one representative vowel of the plurality of vowels. 29. The method of claim 28,
The output unit includes:
Identifying a first candidate word corresponding to a vowel tuple that matches a left vowel combination that is associated with a second syllable based on a beginning syllable of the query by referring to the vowel tuple inverse array database;
Identifying a second candidate word corresponding to a vowel triple matching a right vowel combination that is associated with a third syllable based on the end syllable of the query by referring to the vowel triple inverse array database;
Identifying a word that is common to the first candidate word and the second candidate word as a third candidate word,
Identifying a fourth candidate word corresponding to a vowel triple matching a left vowel combination that is associated with a third syllable based on a beginning syllable of the query by referring to the vowel triple inverse array database;
Identifying a fifth candidate word corresponding to a vowel tuple that matches the right vowel combination associated with a second syllable based on the end syllable of the query by referring to the vowel tuple inverse array database;
Identifying a word common to the fourth candidate word and the fifth candidate word as a sixth candidate word,
And a set-based search apparatus outputting the third candidate word and the sixth candidate word as the search word. 29. The method of claim 28,
When the output search word is plural,
The output unit includes:
And a set-based retrieval device for outputting the first syllable constituting the search word and the second syllable constituting the query word in a higher order. 29. The method of claim 28,
When the output search word is plural,
The output unit includes:
And a set-based retrieval device for the first syllable constituting the search word and the second syllable constituting the query word to be given priority as the number of syllables of which the initial and the neutral match or the neutral and the final match are greater. 29. The method of claim 28,
When the output search word is plural,
The output unit includes:
A first step of increasing the matching score by L according to the number of matches between the first syllable constituting the search word and the second syllable constituting the query word, and the syllable associated with the increase of the matching score in the first step. A second step of increasing the matching score by M according to the number of syllables in which the first and the second syllables match the initial and the neutral or the neutral and the finality of the first syllable and the second syllable, except for the first syllable; In step 2, a third step of decreasing the matching score by N according to the number of the first syllables remaining except for the syllables related to the increase of the matching score is performed, and prioritizing the search word having a relatively high matching score. Set-based search device to output. 34. The method of claim 33,
And L is greater than M and M is greater than N. 2.

Images