JPH0353378A - Name retrieving system for retrieval of family name of same-pronunciation/different-character and different-character/same-pronunciation - Google Patents

Abstract

PURPOSE:To improve the name retrieving efficiency by retrieving the names with a family name using the familiar KANJI (Chinese characters) accustomed to use as a key and also retrieving simultaneously the names of the same- pronunciation/different-character and the different-character/same-pronunciation when a name is retrieved with its family name as a key. CONSTITUTION:When a KANJI family name 'NAKATANI' is inputted via an input device 1, the name 'NAKATANI' is retrieved out of an index file 2 of a KANJI family name/KANA (Japanese syllabary) reading file. The 'NAKATANI' retrieved out of the file 2 has two different reading ways which contain pointers A and B respectively. Thus two KANA reading ways 'NAKATANI' and 'NAKAYA' are written in the entry of a KANJI family/ KANA reading file 3 equivalent to pointers A and B against the KANJI 'NAKATANI'. Then four KANJI names 'NAKATANI TARO', 'NAKATANI JIRO', 'NAKAYA ICHIRO', and 'NAKAYA JIRO' can be retrieved just by inputting the KANJI 'NAKATANI' via the device 1. As a result, the misretrieving frequency is reduced and the retrieving efficiency is improved.

Description

[Detailed Description of the Invention] [Summary] Regarding a name search method in which a surname in kanji is input from an input device and a surname with homophones and homonyms is searched, when a surname in kanji is input on the input device, a specified The purpose of this is to provide a name search method that can simultaneously extract not only names in kanji or kana, but also names with homophone and allograph surnames. , an index file for the name file that searches the original name file using the kana reading of the surname as a key, and an index file for the name file that searches the original name file using the kana reading of the surname as a key, and a surname that extracts only the surname in kanji from the original name file and pairs it with the kana reading. A kanji surname/kana reading file that is stored so that pairs with the surname reading do not overlap, and an index file for the kanji surname/kana reading file for searching the kanji surname/kana reading file using the kanji surname as a key. The structure is equipped with the following.

[Industrial application field]

The present invention inputs a kanji surname from an input device and inputs the homonym (
The following describes a name search method that searches for surnames that have the same kana reading but different kanji (hereinafter referred to as surnames with the same kana reading but different kanji) and alloji homophones (hereinafter refers to surnames with different kanji but the same kana reading).

[Conventional technology]

In a conventional name search, for example, if you enter "Nakaya" in kana using the key on the input device, "Ichiro Nakatani" (pronounced in kana as Nakaya) and "Jiro Nakaya" (pronounced in kana as Nakaya) will be displayed. etc. were extracted.

Furthermore, when the kanji character ``Nakatani'' was entered using the keys on the input device, ``Ichiro Nakatani'' (pronounced in kana as Nakaya) and ``Taro Nakatani'' (pronounced in kana as Nakatani) were extracted.

[Problem to be solved by the invention]

However, when the input device key was used to input ``Nakaya'' in kana, ``Nakatani'' which is a homophone and has a kana reading (such as Taro Nakatani and Jiro Nakatani) was not extracted.

Furthermore, when the kanji character "Nakatani" was entered using the keys on the input device, characters such as Jiro Nakatani and Jiro Nakaya, which also have the same sound but different kanji characters, were not extracted.

As described above, when searching for names with homophones and allophones in the conventional example, not only does the search efficiency deteriorate, but there is a risk of omissions in the search.

The present invention is intended to solve the above problems.
The purpose is to provide a name search method that can simultaneously extract not only the specified kanji or kana name but also names with homophones and homonyms when a surname in kanji is input into an input device. be.

[Means to solve the problem]

FIG. 1 is a diagram explaining the principle of the present invention. In Figure 1,
Enter the name to be searched in kanji using the input device l. In the RAM of the search device (not shown), various files for search are stored. For example, in the index file 2 of the Kanji surname/kana reading file, kanji surnames with homophones and homonyms are stored, and pointers are attached to the kanji surnames with homonyms and homonyms. The Kanji surname/kana reading file 3 stores homophone and allophone kanji and their kana readings written at the pointer specified in the index file 2 of the kanji surname/kana reading file. And kanji surname・
The kana reading file 3 is stored so that pairs of kanji surnames and kana readings do not overlap. In the index file 4 of the name file, surnames with kanji pronunciations corresponding to the kana pronunciations added to the kanji in the kanji surname/kana pronunciation file 3 are filed in kana, and at the same time, pointers are attached to these names. The name file 5 stores the original name corresponding to the kana reading of the last name.

All search results 6 are displayed on the display device, and a desired name is extracted from among them.

[For production]

First, it is assumed that the kanji surname "Nakatani" is input from the input device 1. In this case, "Nakatani" is searched from the index file 2 of the kanji surname/kana reading file. "Nakatani" searched from index file 2 of the kanji surname/kana reading file,
There are two readings, each with a pointer ASB. In the entry of the Kanji surname/kana reading file 3 corresponding to the pointer ASB, "Nakaya" and "Nakatani", which are the kana readings for the above-mentioned "Nakatani", are written.

next,. Using "Nakaya" in the Kanji surname/kana reading file 3 as a key, "Nakaya" is searched from the index file 4 of the name file. There are two types of "Nakaya", and pointers ■ and ■ are attached to each. In the entries in the name file 5 corresponding to the pointers (2) and (2), the same names as the keys, ie, "Ichiro Nakatani" and "Jiro Nakaya" are written. Therefore, these two names are displayed on the display device as search results.

Also, using "Nakatani" in the Kanji surname/kana reading file 3 as a key, "Nakatani" is searched from the index file 4 of the name file. There are two types of "Nakatani", and pointers ■ and ■ are attached to each. The entries in name file 5 corresponding to these pointers ■ and ■ are as follows:
The same names as those on the key, ie, "Taro Nakatani" and "Jiro Nakatani" are written on it. Therefore, these two names are displayed on the display device as search results.

As mentioned above, by simply entering "Nakatani" from the input device L, you can search for the four names "Taro Nakatani,""JiroNakatani,""IchiroNakatani," and "Jiro Nakaya" as search results, and avoid omissions in your search. This will reduce the number of search results and increase search efficiency.

〔Example〕

An embodiment of the present invention will be described with reference to FIG. 2 and FIGS. 3(A) and 3(B). FIG. 2 is a data processing block diagram in the present invention, and FIGS. 3(a) and 3(b) are flowcharts of an embodiment of name search in the present invention.

In FIG. 2, numeral 10 is an input device, for example, a keyboard on which a name can be entered. Reference numeral 11 denotes a control device such as a CPU that operates the system to search for the name input to the input device 10.

Reference numeral 12 denotes a display device, such as a CRT display that can display the input results and search results of the input device 10. Reference numeral 13 denotes a storage device consisting of, for example, ROM and RAM, in which a control program for operating the control device 11 to search for names is stored.

Reference numeral 14 denotes a file control device, which is a control device for searching for a desired name from a file, which will be described later, according to the program in the storage device 13. Further, files 15 to 18 are composed of a storage device capable of transferring data to and from the RAM in the storage device 13, and correspond to each file shown in FIG. That is, the index file 15 of the kanji surname/kana reading file is 2 shown in Figure 1, the kanji surname/kana reading file 16 is 3, the index file 17 of the name file is 4, and the name file 18 is 5. Each corresponds to the other. Candidate file 19 is
This is a buffer memory that temporarily stores candidates retrieved during a name search. The contents of each of these files are transferred to the RAM of the storage device 13 and then processed.
For the sake of simplicity, the following description assumes that the file control device 14 accesses each file as needed.

Next, the name search procedure will be explained according to the flowcharts in FIGS. 3(a) and 3(b).

First, a kanji surname is read in using the input device 10 (FIG. 2) (step 2).

Using the kanji surname read by the input device 10 as a key, the index file 15 (FIG. 2) of the kanji surname/kana reading file is searched (step 2).

It is checked whether there is a record matching the kanji surname specified by the input device 10 in the index file 15 of the kanji surname/kana reading file (step 2).

If a record matching the input kanji surname exists in the index file 15 of the kanji surname/kana reading file, that record is extracted (step 2).

Then, the candidate file 19 ( Figure 2)
(Step ■).

In step (2), if there is no record matching the kanji surname specified on the input device 10 in the index file 15 of the kanji surname/kana reading file, the display screen of the display device 12 (FIG. 2) , displays the message "There are no records corresponding to the specified kanji surname" (step ■).

Furthermore, it is checked whether the record following the record extracted from the index file 15 of the kanji surname/kana reading file matches the kanji surname designated from the input device 10.

This process is repeated as long as there is a designated kanji surname, the record is extracted, and the kana reading of the kanji surname corresponding to the pointer attached to the record is written to the candidate file 19 (step 2).

If there is no longer a record that matches the kanji surname specified by the input device 10, the first record of the candidate file 19 is read (step 2).

The index file 17 (FIG. 2) of the name file is searched using the kana reading read from the candidate file 19 as a key (step 2).

The contents of the records extracted from the index file 17 of the name file are displayed on the display device 12 (step 2).

It is checked whether the record following the record extracted from the index file 17 of the name file matches the kana reading specified in the kanji surname/kana reading file 16 (step 2).

If the name matches the kana reading specified in the kanji surname/kana reading file 16, the name is extracted from the name file 18 (FIG. 2) according to the pointer attached to the record (step 0).

Then, return to step ■ and display the contents of the record on display device 1.
Display on 2.

If the index file I7 of the name file does not contain the contents of the record specified in the kanji surname/kana reading file 16, it is checked whether all the records in the candidate file 19 have been read. If all records of the candidate file 19 have been read, the search ends (step 0>).

If the record of the candidate file 19 has not been read yet, the next record of the candidate file 19 is read and the process returns to step ■ to search the index file 17 of the name file using the kana reading as a key (step ■
).

〔Effect of the invention〕

According to the present invention, when searching for a name using the last name as a key, it is possible to search for the name using the last name in kanji that are familiar to the user as a key, and it is also possible to search for names with homophones and allographs at the same time, resulting in high search efficiency. At the same time, search omissions can be minimized.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of the principle of the present invention, FIG. 2 is a data processing block diagram in the present invention, and FIGS. 3(A) and 3(B) are flowcharts of an embodiment of name search in the present invention. 1...Input device 2, 15...Kanji surname/kana reading file index file 3, 16...Kanji surname/kana reading file 4, 17...
,・Name file index file 5, 18...Name file 10...Input device/control device/display device/storage device/file control device/candidate file

Claims (1)

[Claims] In a name search method in which a name is searched by inputting a surname in kanji, an original name file 5 in which a kana reading is added to the surname of the name and the original name using the kana reading of the surname as a key are provided. Search for file 5 From the name file index file 4 and the original name file 5, extract only the surname from the kanji names and pair it with the kana reading, and the pairing of the kanji surname and the kana reading surname does not overlap. The kanji surname/kana reading file 3 stored as above, and the index file 2 of the kanji surname/kana reading file for searching the kanji surname/kana reading file 3 using the kanji surname as the key. A name search method for searching for surnames with homophones and homonyms.