JP2535629B2 - Input string normalization method of search system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Overview] Regarding an input character string normalization method of a search system for searching a record file such as a database using an input character string as a keyword, it is easy to search by an input character string using an abbreviation that is not spelled correctly. For the purpose of being able to do so, the formal word corresponding to the word cut out from the input character string is searched from the word dictionary, and one or a plurality of normalized character strings are created by the combination of the formal words, and the data is searched.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an input character string normalization method of a search system for searching a record file such as a database using an input character string as a keyword.

In a search system that retrieves necessary data such as an account number from a record file such as a database by using the partner company name etc. as a search keyword by telex telegram etc.,
The input character string such as the company name used as a search keyword must be spelled correctly.

However, abbreviated names are often used in input character strings used as search keywords, and it is possible to register abbreviated names that are expected to be used in addition to official names as keywords. Even so, a system that allows easy data retrieval similar to the official name is desired.

[Prior Art] Conventionally, a system for automatically analyzing and processing telex telegrams and the like from overseas used for bank transactions and the like has been considered.

In such a telegram automatic analysis system, the destination company name spelled in the telegram is used as a search keyword, and a database that records information such as the account number is searched to retrieve the necessary destination data. ing.

[Problems to be Solved by the Invention] However, in a search system that uses an input character string such as a company name as a search keyword, it is rare that the partner company name is input as the official name.
Since it is sufficient if the person in charge of receiving the message can be identified, it is sent in various forms. For example, there are various ones such as those in which English words are omitted, those in which only the proper noun portion of the first word of the company name is input, and those in which the first characters of the words constituting the company name are picked up and arranged.

Furthermore, when writing Japanese in English, there is not only one spelling method, for example, "Tokyo" means "TOKYO" or "TOKI".
Used for both "O".

For such abbreviations of input character strings and different spellings, it is necessary to register abbreviated character strings of various shapes.

However, there is a limit to preparing names that correspond to various inputs of the partner company name, enormous human effort is required to register all possible spellings, and the search dictionary is huge. As a result, there is a problem that the execution area of the program is squeezed and the search efficiency is reduced.

The present invention has been made in view of such conventional problems, and provides an input character string normalization method of a search system capable of easily performing a search by an input character string using an abbreviation that is not correctly spelled. To aim.

[Means for Solving the Problems] FIG. 1 is a diagram illustrating the principle of the present invention.

First, the present invention is directed to a search system that searches the recording file 14 by the search processing means 12 using the input character string from the processed data storage means 10 as a search keyword and outputs the corresponding data.

According to the present invention with respect to such a search system, a word dictionary 16 in which correct spelling formal words corresponding to abbreviations or the like constituting an input character string are stored, and processing data storage means (1
A word cutout unit that cuts out the input character string from 0) into word units, and a formal word obtained when one or more formal words are obtained when the cut out input character string is associated with the word dictionary (16) Is stored in the storage position corresponding to the input character string, and when the official word is not obtained, the storage process of storing the input character string in the storage position corresponding to the input character string and until there are no words to cut out the storage process A normalization consisting of repeated cyclic processing and normalization character string creation processing which creates one or more normalization character strings that can be created by combining stored contents when there are no more words to cut out and outputs to the search means (12). It is composed of a conversion means (18).

[Operation] According to the input character string normalization method of the search system according to the present invention having such a configuration, even an input character string that is not a correctly spelled word such as an abbreviation is abbreviated from the word dictionary as pre-search processing. Search for a formal word corresponding to, a normalized character string is created by a combination of formal words, and the search process is performed using this normalized character string as a keyword, so consider the change in abbreviation or spelling for each word. For example, it is possible to generate a normalized character string containing the correct spelling official name and to effectively perform data search.

[Embodiment] FIG. 2 is a configuration diagram of an embodiment showing one embodiment of the present invention.

In FIG. 2, reference numeral 10 is a data file to be searched, in which data such as telex and telegram are stored by online processing or badge processing. A host computer 20 has the function of the character string normalization processing unit 18 and the function of the search processing unit 12 according to the present invention. A word dictionary file 16 is provided for the character string normalization processing unit 18 of the host computer 20, and a search database 14 is provided for the search processing unit 12. Further, the search result by the search processing unit 12 of the host computer 20 is output to the output device 22 such as a CRT or a printer.

The host computer 20 retrieves the telex and telegram to be processed from the search target data file 10,
An input character string representing the partner company name included in the message is given to the character string normalization processing unit 18, and the abbreviation used in the input character string is converted to a correctly spelled word by referring to the word dictionary file 16 for normalization. Generate a string.

The word dictionary file 16 used for the normalization processing in the character string normalization processing unit 18 stores the correct spelling formal words corresponding to the abbreviations and the like that form the input character string. That is, it is conceivable that the company name, which is the destination company name of the telex and the electronic message, is composed of a business type, a product to be handled, place name information and the like, and a proper noun. So word dictionary file
The words other than proper nouns that compose the company name are registered in 16 along with their abbreviations. For example, the regular spelling "BAN
Since “BK”, “BNK”, “GINKO”, etc. are used as abbreviations for “K”, registration is performed so that the correct spelling of the official word can be searched for each name word.

The outline of the normalization processing by the character string normalization processing unit 18 is as follows.

First, the input character string obtained from the search target data file 10 is word-divided into word units. Next, the word dictionary file 16 is searched for each cut-out word or the like, and one or more corresponding official words are searched. Finally, one or more normalized character strings are created from the combination of formal words obtained from the word dictionary file 16 and passed to the search processing unit 12, and the search processing unit 12 searches using the normalized character strings as keywords. The database 14 is searched. That is, the character string normalization process of the present invention is performed as a pre-process of the input character string such as the partner company name used as a keyword in the search processing unit 12.

Next, the processing operation of the character string normalization processing unit 18 of FIG. 2 will be described with reference to FIGS. 3A and 3B.

In FIG. 3A, first, in step S1 (hereinafter, step is omitted), the input character string obtained from the search target data file 10 is divided by the delimiter and set in the input word group. For example, if the input character string is "NIPPON TEL + TEL" as shown in Fig. 4, the input character string is divided into four parts by the delimiter as the space of each word, and each word is stored in the input word group. .

Next, in S2, the number of input word groups is set to the number N of input word groups. In the case of FIG. 4, the number of input word groups N = 4. Then S3
Then, it is determined whether or not the number of input word groups N = 0, that is, whether or not the normalization processing is completed. First, since the number of input word groups N = 4, the process proceeds to S4.

In S4, the word at the position where the number of input word groups N = 4 in the input word group, for example, the fourth word “NIPPON” from the right in FIG. 4 is set as the search word, and then in S5, the search word “ The word dictionary file 16 is searched using “NIPPON” as a keyword. Search term "NIPPO
Since "N" is a proper noun, it is not registered in the word dictionary file 16 in this embodiment.
From S7, the data corresponding to the search word "NIPPON" is acquired as it is, and in the next S8, the data "NIPPON" acquired in S7 at the position of the input word group index 1 in the word storage area as shown in FIG. To store.

Then, in S9, the number N of input word groups is set to N = 4-1 = 3, and S3 is set.
Returning to S4 again via, the word at the position where the number of input word groups N = 3 in the input word group, that is, the third word “TEL” from the right in FIG. Search the word dictionary file 16 with the search word "TEL" as a keyword. By searching this word dictionary file 16, for the search word "TEL",
You can get two official words "TELEPHONE" and "TELEGRAM". In S6, since there is a word that matches the search word in the word dictionary file, the process proceeds to S10, and the two searches searched in S5 at the position of the input word group index 2 in the word storage area shown in FIG. The word is stored as shown.

For the number of input word groups N = 2, two words "AND" and "PLUS" are obtained by searching the word dictionary file 16 with the second word "PLUS" from the right in FIG. 4 as the search word. Even in the last process in which the number of input word groups is stored as shown in FIG. 5 as shown in FIG. Two words are searched, and the search word is stored as shown at the position of the input word group number index 4 in FIG.

When the search process of the word dictionary file 16 is completed for all delimiters of the input string,
The number of input word groups N = 0 is determined in S3, and the process proceeds to S11 in FIG. 3B.

In S11, the number of input word groups N = 4 is set in the input word number index, then the process proceeds to S12, and one word is sequentially taken out from the position of index 4 in the word storage area shown in FIG. The index is decremented by 1 in S13, the index is 0 in S14, that is, it is determined whether four words have been extracted. If four words have not been extracted yet, the process returns to S12 and the next index 3 word is extracted. Is taken out, and thereafter, the word taking out is repeated in order until the index = 0. When the extraction of the four words is completed, the index becomes 0, so the process proceeds to S15, and the normalized character string composed of the combination of the four words extracted from the word storage area of FIG. 5 is stored in the character string storage area. Subsequently, in S16, it is checked whether or not all the patterns have been stored in the character string storage area. If not stored, the process returns to S11 and the number of input word groups N = 4 is set again in the index to extract the next pattern. Do.

When it is determined in S16 that all the patterns are stored in the character string storage area, the process proceeds to S17, the stored data in the character string storage area is delivered to the search processing unit 12, and the search processing unit 12 causes the character string normalization processing unit 18 to operate. The search process of the search database 14 is executed using the normalized character string in the character string storage area delivered from the as a keyword.

FIG. 6 shows the storage state of the normalized character string created from the storage data of the word storage area shown in FIG. 5 in the processing of S11 to S17 in FIG. 3B.

That is, as shown in FIG. 5, for the input character strings shown in FIG. 4, since the dictionary files are searched for the correctly spelled words indicated by indexes 1 to 4 shown in FIG. According to 8 in Fig. 6
A pattern of two normalized strings is generated.

The eight patterns generated as shown in FIG. 6 are sequentially used as keywords in the search processing unit 12 to search the search database 14, and if the character string is a correct normalized character string, the search result can be obtained. In the case of FIG. 6, since the pattern is a formal name, the corresponding search result can be obtained by the search process using the normalized character string of as a keyword.

[Effect of the Invention] As described above, according to the present invention, even if only an input character string with a correct spelling is effective as a search keyword for a recording file, an input character string using an abbreviation is correct by normalization processing. By converting to a spelled character string, it is possible to obtain the corresponding search result even with an input character string using an abbreviation, reduce search errors, reduce unnecessary operations, and improve search performance. it can.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is an explanatory view of the principle of the present invention; FIG. 2 is a configuration diagram of an embodiment of the present invention; FIGS. 3A and 3B are character string normalization processing flow charts of the present invention; 4 is an explanatory diagram of an input character string; FIG. 5 is an explanatory diagram of an input word number index and search data storage; FIG. 6 is an explanatory diagram of a character string storage area. In the figure, 10: processed data storage means (search target data file) 12: search processing means 14: database (search database) 16: word dictionary (word dictionary file) 18: character string normalization means 20: host computer 22: output apparatus

Claims (1)

(57) [Claims] 1. A search system for searching a record database (14) by a search processing means (12) using a character string from a processed data storage means (10) as a search keyword to output corresponding data A word dictionary (16) storing correct spelling formal words corresponding to abbreviations forming columns; word cutting processing for cutting an input character string into word units from the processing data storage means (10), and the cut input When one or more formal words are obtained when the character string is associated with the word dictionary (16), all the obtained formal words are stored in the storage positions corresponding to the input character string to obtain the formal word. If not, a storage process of storing the input character string in a storage position corresponding to the input character string, and a cyclic process of repeating the storage process until there are no more words to cut out,
A normalizing means comprising a normalizing character string creating process for creating one or a plurality of normalizing character strings that can be created by combining the stored contents when there are no more words to be cut out, and outputting the normalizing character string to the searching means (12). (18) and; are provided, and the input string normalization method of the search system is characterized.