JP2018101224A - Searching apparatus, searching method, and program - Google Patents

Abstract

Even when a first type character string corresponding to desired information cannot be specified, desired information can be easily searched. In a search device for performing a search on a database based on an input of a first type character string consisting of one or more characters of a first character type, a first type character for each first type character string. A data storage unit 104 that stores a second type character string made up of one or more characters of the second character type associated with the column, an input unit 101 that receives an input of the first type character string, and an input unit 101 The input first type character string, which is the first type character string input by the above, is divided for each element, and for each divided element, one or more second type character strings corresponding to the divided element are stored in the data storage unit 104. And a conversion / search processing unit 103 that searches the database 105 using the acquired second type character string and acquires the search result, and a display unit 102 that outputs the search result. [Selection] Figure 1

Description

  The present invention relates to a technique for searching for information based on an input character string.

  When searching a database consisting of information on kana characters by entering a character string of alphabetic characters (which may also be called roman characters), the alphabetic character string is converted to a kana character string, Search the database using character strings.

  In the case of converting an alphabet character string into a kana character string, the conventional technique uses a one-to-one correspondence between the kana character and the alphabet character string to convert the alphabet character string into a kana character. Convert to column. As a prior art document related to such a conversion technique, for example, there is Patent Document 1.

Japanese Patent No. 4001872

  In general, a kana character string has a plurality of alphabet characters. For example, “onishi” can be written as onishi, oonishi, ohnisi, etc., and all of these are strings of correct alphabetic characters (Roman characters) indicating “onishi”.

  However, in the above-described conversion method in the prior art, “o” is converted to “o”, and “on” is not converted to “o”. Therefore, for example, even if onishi is entered and a database composed of kana character information is searched, the expected search result “Oonishi” cannot be obtained. Therefore, in the prior art, when searching a database composed of kana character information by inputting an alphabetic character string, the user can use the alphabetic character string according to the format registered on the system side. It was necessary to input, and the search with an ambiguous character string was not possible, and convenience was low. In particular, in the case of a person name, the notation tends to vary, and even in the case of the same sex, a character string of a different alphabet character is often registered, and the search is not easy.

  The problems as described above are not limited to conversion from alphabetic characters to kana characters, but are problems that may occur in conversions related to various character types.

  The present invention has been made in view of the above points, and a first type character string that is a character string expressed in characters of the first character type is a second character string that is expressed in characters of the second character type. In the technique of converting to a seed character string and searching for desired information using the second kind character string, even when the first kind character string corresponding to the desired information cannot be specified, it is easy to search for the desired information It is an object to provide a technique that can be performed in the future.

According to the disclosed technology, a search device that performs a search on a database based on an input of a first type character string composed of one or more characters of a first character type,
For each first type character string, a storage unit that stores a second type character string made up of one or more characters of the second character type associated with the first type character string;
An input unit for receiving input of the first type character string;
The input first type character string that is the first type character string input by the input unit is divided for each element, and for each divided element, one or more second type character strings corresponding to the divided element are A processing unit that acquires from the storage unit, searches the database using the acquired second type character string, and acquires a search result;
And an output unit that outputs the search result.

  According to the technology of the disclosure, the first type character string that is a character string expressed in characters of the first character type is converted into the second type character string that is a character string expressed in characters of the second character type, In a technique for searching for desired information using a second type character string, a technique that makes it possible to easily search for desired information even when the first type character string corresponding to the desired information cannot be specified. Is provided.

It is a functional lineblock diagram of search device 100 in an embodiment of the invention. 2 is a hardware configuration diagram of the search device 100. FIG. 4 is a flowchart showing an overall flow of processing of the search device 100. It is a figure which shows the example of a process with respect to an alphabet character string. It is a figure which shows the example of a display. It is a flowchart which shows the detailed process example of step S1. It is a figure which shows the example of the table referred by step S1. It is a flowchart which shows the detailed process example of step S2. It is a figure which shows the example of the table referred by step S2.

  Hereinafter, an embodiment (this embodiment) of the present invention will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

  In the following, an example will be described in which a character string expressed in alphabetic characters is converted into a character string expressed in Kana characters, and a database is searched using the converted character string, but this is only an example. . The character type of the conversion source may be a character type other than alphabetic characters, and the character type of the conversion destination may be a character type other than kana characters. The kana characters may be hiragana or katakana. Below, the case where hiragana is used is shown as an example.

  Hereinafter, a character string expressed in alphabet characters is called an alphabet character string, and a character string expressed in kana characters is called a kana character string. The alphabet character string is a character string made up of one or more alphabet characters, and the kana character string is a character string made up of one or more kana characters. In addition, you may paraphrase the alphabetic character and alphabetic character string in this Embodiment as a Roman character and a Roman character string, respectively.

  In addition, the application field of the search in the following embodiment is not limited to a specific field, but as an example, it can be applied to a search for a personal name database in a predetermined organization used for a seating chart or the like.

(Device configuration)
FIG. 1 shows a functional configuration diagram of the search device 100 according to the present embodiment. As shown in FIG. 1, the search device 100 according to the present embodiment includes an input unit 101, a display unit 102, a conversion / search processing unit 103, and a data storage unit 104. Further, as shown in FIG. 1, a database 105 storing information to be searched is connected to the search apparatus 100 via a network. The database 105 may be provided inside the search device 100. The data storage unit 104 may be provided outside the search device 100 and connected to the search device 100 via a network. Even if the data storage unit 104 is connected to the search device 100 via a network, the configuration including the search device 100 and the data storage unit 104 may be referred to as a “search device”.

  The input unit 101 receives an input of an alphabet character string from the user. The display unit 102 displays the alphabet character string input by the input unit 101 and also displays the search result by the conversion / search processing unit 103. The display unit 102 is an example of an output unit that outputs search results. Instead of the display unit 102, an output unit that does not perform display may be provided. In this case, for example, the search result output from the output unit is transmitted to another device, and the search result is displayed on the other device. For example, the search device 100 operates as a web server or an application server, and a character string input from a user terminal (client) connected to the network is input from the input unit 101 to the search device 100, and the search result is output as described above. The configuration may be returned from the unit to the user terminal.

  The conversion / search processing unit 103 converts the alphabet character string input from the input unit 101 into a kana character string, and performs a search on the database 105 using the kana character string. The data storage unit 104 stores various tables that are referred to when converting an alphabet character string into a kana character string.

  The database 105 is a database that stores search target information composed of kana character strings. In this embodiment, the search process is performed with kana character strings, but information other than kana character examples is also stored in the database 105, and the search apparatus 100 also displays information other than kana character strings as search results. Is possible. For example, the name of a person can be displayed in kanji or alphabet, or an e-mail address, telephone number, or the like can be displayed. The database 105 may be a database having only kana character information (no alphabet characters).

  The input unit 101 or the conversion / search processing unit 103 has an existing kana-kanji conversion function, converts the input alphabet character string into a kana character string or kanji, and the converted kana character string or kanji is displayed on the display unit 102. May be displayed. In addition, when an existing Kana-Kanji conversion function is provided, characters that can be converted by the Kana-Kanji conversion function, such as “Ut” and “Stone t”, and non-converted alphabet characters may be displayed together. Good.

  Even when a kana character string or kanji is displayed together with unconverted alphabet characters by the existing kana-kanji conversion function, the search processing described below can be executed as it is.

<Hardware Configuration Example of Search Device 100>
The search apparatus 100 according to the present embodiment can be realized, for example, by causing one or a plurality of computers to execute a program describing the processing content described in the present embodiment. That is, the function of the search device 100 is realized by executing a program corresponding to the process executed by the search device 100 using hardware resources such as a CPU, a memory, and a hard disk built in the computer. Is possible. The above-mentioned program can be recorded on a computer-readable recording medium (portable memory or the like), stored, or distributed. It is also possible to provide the program through a network such as the Internet or electronic mail.

  FIG. 2 is a diagram illustrating an example of a hardware configuration of the search device 100 when the search device 100 is realized by a computer. The search device 100 illustrated in FIG. 2 includes a drive device 150, an auxiliary storage device 152, a memory device 153, a CPU 154, an interface device 155, a display device 156, an input device 157, and the like that are mutually connected by a bus B.

  A program that realizes processing in the search device 100 is provided by a recording medium 151 such as a CD-ROM or a memory card, for example. When the recording medium 151 storing the program is set in the drive device 150, the program is installed from the recording medium 151 into the auxiliary storage device 152 via the drive device 150. However, the program does not necessarily have to be installed from the recording medium 151, and may be downloaded from another computer via a network. The auxiliary storage device 152 stores the installed program and also stores necessary files and data.

  The memory device 153 reads and stores the program from the auxiliary storage device 152 when there is an instruction to start the program. The CPU 154 (processor) implements functions related to the device according to a program stored in the memory device 153. The interface device 155 is used as an interface for connecting to a network. The display device 156 displays a GUI (Graphical User Interface) or the like by a program. The input device 157 includes a keyboard and mouse, buttons, a touch panel, and the like, and is used to input various operation instructions.

(Flow of processing executed by search device 100)
Next, the flow of processing executed by the search device 100 having the configuration shown in FIG. 1 will be described along the procedure of the flowchart of FIG. The following steps S1 and S2 are outlines, and detailed processing will be described later.

  In step S1, the alphabetic character string input from the input unit 101 is passed to the conversion / search processing unit 103, and the conversion / search processing unit 103 ensures that the alphabetic character string does not straddle a kana character in Roman alphabet notation. Divide into unit alphabet strings. As an example, when “onis” shown in FIG. 4A is input, it is divided into three alphabetic character strings “o”, “ni”, and “s”. If it is divided into “on” and “is”, the kana character “ni” straddles these two elements, so that “on” and “is” are not divided.

  In step S2, the conversion / search processing unit 103 analyzes each alphabet character string divided in step S1 and refers to the conversion table stored in the data storage unit 104, thereby converting the alphabet character string into Replace with one or more kana strings. In the example of FIG. 4, as shown in (b), “o” is replaced with “o”, “o”, and “o”, “ni” is replaced with “ni”, and “s” is Replaced by “sa”, “shi”, “su”, “se”, “so”. Further, regarding “n” and “i” constituting “ni”, “n” is replaced with “n”, and “i” is replaced with “i”. These replaced character strings are stored in a candidate array (memory area).

  In step S3, the conversion / search processing unit 103 converts the kana character string set obtained in step S2 into a regular expression so that all possible combinations of kana character strings can be searched. In the example of FIG. 4, conversion is performed as shown in (c). In the regular expression shown in (c), “^” indicates the beginning of a line, and “|” indicates “or”.

  In step S4, the conversion / search processing unit 103 searches the database 105 using the regular expression created in step S3.

  In step S5, the conversion / search processing unit 103 passes the search results obtained from the database 105 to the display unit 102, and the display unit 102 displays the search results. In the case of the regular expression shown in FIG. 4C, for example, “Onishi”, “Onishi”, “Ounsai” and the like are matched by the search and displayed on the display unit 102. The conversion / search processing unit 103 may acquire only the matched kana character string from the database 105 and display it on the display unit 102, or information associated with the matched kana character string (eg, kanji name, romaji) Name etc.) may also be acquired and displayed on the display unit 102.

  FIG. 5 shows an example of the display screen. The example shown in FIG. 5 is an example when “ot” is input. In this case, “Ota”, “Otsuki”, and “Ochi” are matched by the conversion / search processing by the conversion / search processing unit 103, and information corresponding to these matches is displayed.

  In the above example, the case where “onis”, “ot”, or the like is input is described. However, for each input of one alphabet, the search apparatus 100 inputs the one alphabetic character and the previous one. The above-described processing is executed on the character string composed of the alphabet characters, and the search result is displayed sequentially for each input of one alphabet. In other words, an ambiguous search + incremental search is realized in which alphabetical character strings that are not sure whether or not they match desired information are sequentially input to narrow down the search results.

  For example, when "o" is entered, information of four people, Oishi, Ogata, Oguri, and Ue, are displayed as search results, and then when "h" is entered, search for "oh" As a result, the information of two persons, Oishi and Oue, are displayed as search results. When n is entered next time, the information of one person of Oishi, as search results, is retrieved as the search result for “ohn”. Is displayed. Also, for example, when “o” is entered, information on the four names of Oishi, Ogata, Oguri, and Oue are displayed as search results, and then when “u” is entered, “ou” As a search result, information of one person is displayed as the search result. Also, for example, when “o” is entered, the information of the four names of Oishi, Ogata, Oguri, and Oue are displayed as search results, and then when “g” is entered, “og” As a search result, information of two persons, Ogata and Amaguri, is displayed as a search result.

  Also, for example, when the information of four persons, Oishi, Ogata, Oguri, and Ue, is displayed as a search result when “o” is entered, a specific person is entered from the input unit 101. When information indicating confirmation is input, the conversion / search processing unit 103 may select the specific person's information and pass it to the display unit 102, and the display unit 102 may display the information. The information indicating that a specific person is confirmed may be, for example, selecting one conversion candidate (for example, Onishi) from a plurality of conversion candidates displayed by the existing kana-kanji conversion function.

  In other words, the technique according to the present embodiment displays a plurality of candidates for information (for example, the name of a person) that the user wants to search just by entering one letter of the alphabet, and narrows down the candidate for the name each time one more character is added. Can be displayed. As a result, it is possible to quickly search for a person name to be searched, and it is possible to search even when an ambiguous character is input. Also, when combined with an existing Kana-Kanji search, unlike the exact match search, the search result can be displayed even when unconverted Roman characters being input are mixed in the character string.

(Detailed processing example of step S1)
Hereinafter, the alphabet character string dividing process executed by the conversion / search processing unit 103 in step S1 will be described in more detail with reference to the flowchart of FIG.

  In step S101, a start index (start index) and an end index (end index) are initialized to zero. In step S102, the end index is incremented by one. In the example of “onis” in FIG. 4A, the end index indicates “o” at this point. Hereinafter, description will be made using an example of “onis” in FIG.

  In step S103, the character ("o") indicated by the end index is analyzed. In step S104, it is determined whether the character ("o") indicated by the end index is the last character of the character string ("onis"). If Yes, the process proceeds to step S105. If No, the process proceeds to step S106. move on.

  In step S105, the substring is added to the character set and the process ends. A substring is a character string from the character after the start index to the character at the end index. For example, in the example of FIG. 4A, when the start index is 0 and the end index is 1, the substring is “o”, and when the start index is 1 and the end index is 3, the substring is “ ni ". The character set is a name of a character string to be processed in step S2. In the case of proceeding to step S105, in the example of FIG. 4A, the substring “s” is added to the character set. In FIG. 6, “set” in S101 and S109 means setting, and “set” in S105, S106, S107 and S108 means a character set.

  In step S106, it is determined whether or not the substring satisfies a character set end condition. FIG. 7 shows an example of the character set end condition. The character set end condition table is stored in the data storage unit 104 and is referred to by the conversion / search processing unit 103. Note that “char” in the first line in FIG. 7 is an abbreviation for “character”.

  In the case of the entry on the first line of the set end condition shown in FIG. 7, if the character in the end index corresponds to any of the five characters of aieueo and the next character is not h, the determination in step S106 is Yes. It becomes. Other entries are also determined according to the entry description. increment startIndex is a value used in step S109.

  In step S107 when the determination in S106 is Yes, it is determined to add a substring to the set, and in step S108, the substring from the start index to the end index (from the character indicated by the index next to the start index, Character string up to the character pointed to by the end index) is added to the set. In step S109, the start index is set to “end index + increment startIndex”. For example, when the start index is 0 and the end index is 1 and the increment startIndex is 0, the start index is 1. After step S109, the process proceeds to step S102.

  For example, in the example of FIG. 4A, it is assumed that at step S109, the start index is 1 and the end index is 1, and the process proceeds to step S102. In this case, the character n indicated by the end index 2 is No in the determination in step S106, and the process returns to step S102 again. At the time of step S102, the start index is 1 and the end index is 3. In this case, the determination in step S106 is Yes, and “ni” is obtained.

(Detailed processing example of step S2)
Next, the process of replacing the alphabetic character string with the kana character string executed by the conversion / search processing unit 103 in step S2 described above will be described in more detail with reference to the flowchart of FIG. In the present embodiment, the process shown in FIG. 8 is performed for each alphabetic character string (for each character set) divided in step S1. In the following description, the example of FIG. 4 is used as appropriate. In the following example, the tables shown in FIGS. 9A to 9C are referred to as appropriate. These tables are tables stored in the data storage unit 104.

  In step S201, it is determined whether the character set is empty. If the determination in step S201 is Yes, the process proceeds to step S216, and the replaced character string (candidate character string) is returned. If the determination in step S201 is No, the process proceeds to step S202. In this example, processing is recursively executed for each subset (eg, “n” and “i” in “ni”) in an alphabetic character string that constitutes a certain character set. If there is no unprocessed subset in the character set, S201 is Yes.

  Here, first, it is assumed that the character set “o” is processed. In step S202, the length of the subset is set to 1, and the subset is acquired in step S203. Here, since the character set is only one character “o”, this is acquired as a subset.

  In step S204, it is determined whether or not the subset satisfies a condition in a special case table. FIG. 9C shows an example of the special case table. The example of the special case table shown here shows a rule that “t” is added when the target subset starts with continuous consonants (eg, pp).

  If this condition is met, the process proceeds to step S213 to perform processing according to the rule, but “o”, which is an example of the subset here, does not meet the condition, and the process proceeds to step S205.

  In step S205, it is determined whether or not the corresponding subset is the last subset of the last character set (“s” in the example of FIG. 4B). “O” is No, and the process proceeds to step S208.

  In step S208, it is determined whether the subset exists in the conversion table. Since “o” exists as shown in FIG. 9A and becomes “Yes”, the process proceeds to step S207, and the value “Oh, ou” in the conversion table is added to the candidate sequence.

  Next, in steps S209 to S212, the above-described processing is recursively performed on the target character set to perform processing for acquiring candidates.

  However, since the character set “o” here is one character, the process proceeds to step S214. In step S214, the subset length is incremented (that is, set to 2), and the process proceeds to step S215. In step S215, since the subset length (2) is longer than the character set length (1) (that is, longer than the maximum subset length), the process proceeds to step S216 and a candidate is returned. In other words, “Oh, ooh” is returned. In the form of an array, [o, o, o] is returned.

  Next, the process for the character set “ni” is performed. Here, in step S203, the subset “n” is acquired, the process proceeds from step S204 to S205 to S208, and corresponds to the conversion table. In step S207, “n” is added to the candidate sequence. Subsequently, “i” is acquired for the subset “i” by recursive processing (steps S209 to S212), and “i” is added to the candidate sequence. Thereby, a candidate sequence [n, i] is obtained.

  Next, the subset length is set to 2 in step S214, and after step S215, the subset “ni” is acquired in step S203. The process proceeds from step S204 to S205 to step S208 and corresponds to the conversion table. Therefore, in step S207, “ni” is added to the candidate sequence. Thereafter, after steps S214 and S215, in step S216, the candidate sequence [“ni”, [n, i]] is returned.

  Next, processing is performed on the character set “s” (= subset). In this case, “Yes” is determined in step S205, and in step S206, it is determined whether or not the subset “s” is in the “terminus table”. The table is shown in FIG. 9B, and this table also includes information of “conversion table” in FIG. 9A. However, the “terminus table” may not include “conversion table” information. In this case, if the subset matches in either “conversion table” or “terminus table” in step S206, the result is Yes.

  Here, since s exists in the “terminus table”, step S206 is Yes, and in step S207, a value is added to the candidate sequence from the value of the table, and the candidate sequence [sa, shi, shi, se, so] is changed. can get. Thereafter, [S, S, S, S, S] is returned in step S216. As a result, a kana character string for each character set is obtained as shown in FIG.

  As described above, when the last consonant of the input alphabetic character string is registered in the table of FIG. 9B, all the kana characters that have the consonant in their head are acquired when they are written in Roman letters.

  In addition, it is good also as performing the said process of step S2 without performing the division | segmentation of step S1. Although the amount of calculation increases, the same result as that obtained when the division in step S1 is performed is obtained. Even when the division in step S1 is not performed, the process corresponding to the division is executed by the process in step S2.

(Summary of embodiment)
As described above, according to the present embodiment, a search device that performs a search on a database based on an input of a first type character string composed of one or more characters of a first character type, the first type character A storage unit for storing a second type character string made up of one or more characters of the second character type associated with the first type character string for each column, and an input for accepting an input of the first type character string And an input first type character string that is a first type character string input by the input unit, and for each element divided, one or a plurality of second types corresponding to the divided element A character string is acquired from the storage unit, the database is searched using the acquired second type character string, and a processing unit that acquires a search result and an output unit that outputs the search result are provided. A search device is provided.

  The storage unit stores a predetermined character string that can be a terminal character string of the input first-type character string and a plurality of second-type character strings that are associated with the predetermined character string. And when the character string at the end of the input first type character string as one of the delimited elements corresponds to the predetermined character string, the plurality of second type character strings are stored in the storage unit. It is good also as acquiring and using for the said search. For example, the character of the first character type is a Roman character, and the predetermined character string is a consonant. Further, for example, the second character type character is a kana character, and the plurality of second type character strings are kana characters having a consonant in Roman alphabet notation. Further, for example, the first character type character is a Roman character, the second character type character is a kana character, the predetermined character string is a consonant, and the plurality of second type character strings are expressed in Roman letters. A kana character having the consonant.

  The output unit displays all search results obtained by the processing unit, and displays only the specific search results when information indicating that the specific search results are confirmed is input from the input unit. It is good to do. The information indicating that the specific search result is confirmed is, for example, a specific kanji character string candidate from the converted kanji character string candidates displayed by the kana-kanji conversion function based on the input first type character string. This is information indicating that it has been selected.

  For example, the database is a database of personal names in a predetermined organization, and the processing is performed even when a first type character string corresponding to only a part of a specific personal name included in the database is input by the input unit. The unit may obtain a search result including the specific person name. That is, even if the user remembers a person's name, the user can search for information on the person's name.

  Although the present embodiment has been described above, the present invention is not limited to the specific embodiment, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims. Is possible.

100 Search Device 101 Input Unit 102 Display Unit 103 Conversion / Search Processing Unit 104 Data Storage Unit 105 Database 150 Drive Device 151 Recording Medium 152 Auxiliary Storage Device 153 Memory Device 154 CPU
155 Interface device 156 Display device 157 Input device

Claims (8)

A search device for performing a search on a database based on an input of a first type character string made up of one or more characters of a first character type,
For each first type character string, a storage unit that stores a second type character string made up of one or more characters of the second character type associated with the first type character string;
An input unit for receiving input of the first type character string;
The input first type character string that is the first type character string input by the input unit is divided for each element, and for each divided element, one or more second type character strings corresponding to the divided element are A processing unit that acquires from the storage unit, searches the database using the acquired second type character string, and acquires a search result;
An output unit that outputs the search result. The storage unit stores a predetermined character string that can be a terminal character string of the input first type character string, and a plurality of second type character strings associated with the predetermined character string,
The processing unit stores the plurality of second type character strings when the character string at the end of the input first type character string as one of the divided elements corresponds to the predetermined character string. The search device according to claim 1, wherein the search device is acquired from a section and used for the search. The search device according to claim 2, wherein the character of the first character type is a Roman character, and the predetermined character string is a consonant. 4. The search device according to claim 2, wherein the second character type character is a kana character, and the plurality of second type character strings are kana characters having a consonant in Roman alphabet notation. 5. The output unit displays all search results obtained by the processing unit, and displays only the specific search results when information indicating that the specific search results are confirmed is input from the input unit. The search device according to any one of claims 1 to 4, wherein: The database is a database of names of persons in a predetermined organization, and even when a first type character string corresponding to only a part of a specific person name included in the database is input by the input unit, the processing unit The search apparatus according to claim 1, wherein a search result including the specific person name is acquired. A search method executed by a search device that performs a search on a database based on an input of a first type character string made up of one or more characters of a first character type,
The search device includes a storage unit that stores, for each first type character string, a second type character string made up of one or more second character type characters associated with the first type character string,
An input step for receiving an input of the first type character string;
The input first type character string, which is the first type character string input in the input step, is divided for each element, and for each divided element, one or more second type character strings corresponding to the divided element are obtained. A processing step of acquiring from the storage unit, searching the database using the acquired second type character string, and acquiring a search result;
An output step for outputting the search result.   The program for functioning a computer as each part in the search device of any one of Claims 1 thru | or 6.

Images