JP2541944B2 - Sorting substring combination processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Outline] In a device that performs a process such as an anagram that restores characters and words that are arranged in a disjoint order in a correct order by a computer, the problematic lump of characters or partial character strings is detected. Focusing on one solution by processing only the combination with the previous lump of the insertion destination, the combination with the lump after the insertion destination, and the binding of the front and rear lumps at the extracted position when moving ， The problem is dynamically changed, and high-speed join processing is possible.

[Industrial applications]

The present invention relates to a device for performing a process of restoring characters and words arranged in a random order by a computer, and particularly, for processing a word play called anagram by a computer, a high-speed sub-character string The present invention relates to a rearranged partial character string combination processing method that enables combination processing.

In recent years, attention has been paid to using computers for education to enhance learning effects. If so-called anagrams can be processed by a computer, various materials such as Japanese, English, and Chinese can be selected as the material, and it is considered possible to provide a device for learning various languages while having fun.

[Conventional technology]

FIG. 4 shows an operation example of an anagram by a computer for explaining the problem of the present invention.

In this example, "Higinouru" is the problem material. A problem is that the problematic material is rearranged in random units using, for example, a random number. By rearranging the letters or substrings in question by input, and finally obtaining the same sequence of question materials, the correct answer is reached.

The unit of movement for rearrangement is a block of characters or substrings (hereinafter referred to as a chunk), and a character or substring that has been once combined will not be divided thereafter.

The initial state of the problem shown in Fig. 4 is "no", "ru",
"Ga", ".", "Bo", and "hi" are chunks C01 to C06, respectively. Here, chunk "bo",
If there is an instruction to put between "no" and "ru", the problem material and the sequence of questions are collated, and "no", "bo" and "ru" are connected to form chunk C11.

Next, the chunks C11 to C14 are treated as a new problem, and similarly, the joining process is performed on the chunk move instruction.
Finally, when there is one chunk like chunk C31, it is complete.

Conventionally, when character data of a certain length is disassembled in this way, the processing of which part of the original character data corresponds to that part of the character data is LISP or PROLOG, which is most good at symbol processing. It is considered to create a program that performs processing to check the correspondence by all combinations by using the language for artificial intelligence of or by an assembler or the like.

[Problems to be solved by the invention]

When processing anagrams etc. by computer,
If you don't react fast, your interest will drop in half. However, LISP and P
Languages for artificial intelligence such as ROLOG have a problem that the processing speed is slow and a large amount of memory is consumed. Even if an assembler or the like is used, it may not be usable in terms of processing speed. The reason will be described with reference to FIG.

FIG. 5 is a diagram showing an example of the correspondence between question materials and questions.

It is assumed that the problematic material is “Haruogaki, Haruko Mokuru” as shown in FIG. 5 (a). On the other hand, if the sequence of problems in the initial state or in the middle is "Kogarugu, Harumokuru", there are various correspondences between the problem part and the question material. .

In the example of FIG. 5 (b), “ko”, “haru”, “gakite, haru”, “o”, and “mokuru” are associated with the question material as chunks, respectively. On the other hand, as shown in Fig. 5 (c), "ko", "haru", "gakite,",
Correspondence by chunks such as "Haruo" and "Mokuru" is also possible.

In general, as a technique related to artificial intelligence, a method is often adopted in which a solution (candidate) that can be a solution is given as much as possible in the middle of the calculation process, and the correct one is narrowed down. In terms of logic, this gives the correct solution.

Based on such a conventional idea, when the correspondence relationship shown in FIG. 5 is obtained, every correspondence relationship between the problem material and the problem is identified every time the chunk is moved, and the number of chunks becomes the smallest. , It was considered to select chunks by so-called longest match. Therefore, when the problem material becomes long, the processing time by the computer becomes very long.

The present invention solves the above-mentioned problems, and by providing a method in which one effective solution can be obtained without searching all the solutions, the processing time is shortened and the memory usage is reduced. Is intended.

[Means for solving problems]

 FIG. 1 is a block diagram of the principle of the present invention.

In FIG. 1, 10 is a processing unit consisting of a CPU and a memory, 11 is an input device such as a keyboard and mouse, 12 is a display, 13 is a problem material data storage unit for storing character data of problem material, and 14 is a dynamic A problem chunk list storage unit that stores the sequence of changing problem chunks, 15 is a chunk correspondence information storage unit that stores correspondence information between the current problem and the problem material, and 16 is a problem proposal that poses the first problem Section,
17 is a chunk move processing unit that extracts the specified chunk from the problem and then inserts that chunk into the specified move destination, and 18 is a head join process that processes the join between the move chunk and the previous chunk at the insert destination Part, 19 is a tail join processing part that joins the moving chunk and the chunk after the insertion destination, 20 is an intermediate join processing part that joins the chunks before and after the chunk is extracted, and 21 is the final goal Goal judgment part to check whether or not
Represents a result display unit for displaying the result on the display 12.

The question material data storage unit 13 stores the data of the question material provided by the system or the question material previously input by the user. The problem presenting unit 16 partially divides the problem material and divides it into random numbers or the like to create a problem, stores chunk information about the problem in the problem chunk list storage unit 14, and displays the problem on the display 12
To be displayed.

The problem chunk list storage unit 14 stores information about a sequence of dynamically changing problem chunks (hereinafter referred to as Meta-Q). The chunk correspondence information storage unit 15 stores information (hereinafter, referred to as Meta-A) about which part of the problem material each chunk stored in the problem chunk list storage unit 14 corresponds to. It can be considered that Meta-A is composed of pointer information that links Meta-Q and problem material.

The chunk movement processing unit 17 performs processing for extracting one of the chunks in Meta-Q and inserting it at another position in response to the input from the input device 11. For example, assume that chunk C3 is extracted and inserted between chunk C1 and chunk C2.

The head joining processing unit 18 checks the joining of the extracted chunk C3 and the preceding chunk C1 using the chunk correspondence information storage unit 15, and if joining is possible,
The process of combining the chunks into one chunk is executed. In this case, since "D" and "A" are not combined,
It remains as it is.

The tail combination processing unit 19 checks the combination of the extracted chunk C3 and the subsequent chunk C2. In this example, if the combination is checked via the chunk correspondence information storage unit 15, it means that the corresponding sequence of "ABC" is in the problem material data storage unit 13, so that the combination is possible.
The chunk C3 and the chunk C2 stored in the problem chunk list storage unit 14 are combined to create a new chunk C2. In addition, the contents of the chunk correspondence information storage unit 15
Update Meta-A and reconstruct chunks.

The intermediate joining processing unit 20 joins the two chunks C2 and C4 before and after the extracted position. In this example, the new chunk "ABC" is checked as chunk C2. Goal determination unit after processing by the intermediate join processing unit 20
21 determines whether the Meta-Q has the same character sequence as the problematic material.

The result display unit 22 displays the processing results up to that point and outputs a correct answer message when the goal is reached.
If the goal has not been reached, the Meta-Q of the problem chunk list storage unit 14 is treated as a new problem and the process is continued in the same manner.

[Action]

The chunk correspondence relationship information storage unit 15 holds the correspondence information between the chunk in the current question and the correct answer in the question material data storage unit 13, and when a chunk is moved,
Top join processing unit 18, tail join processing unit 19, intermediate join processing unit 20
The join processing is performed only for the position related to the movement. This ensures that only one solution is required.

Therefore, the target solution can be obtained at an extremely high speed as compared with the case where all the correspondences between the problem material or Meta-A and Meta-Q are identified.

〔Example〕

FIG. 2 is a diagram for explaining the processing of one embodiment of the present invention, and FIG. 3 shows an example of the combining processing according to one embodiment of the present invention.

The flow of processing according to an embodiment of the present invention will be described below with reference to FIG. The numbers in the following explanation are
It corresponds to the processes 1 to 3 shown in FIG.

First, a given question material is divided and divided into pieces to create a question, and the chunk information of the question is set in the question chunk list storage unit 14 and the chunk correspondence information storage unit 15, and the display is displayed. Display on 12 and raise the issue.

 Chunk movement information is input from the input device 11.

Check the extraction and insertion positions of the chunk. Then, the extraction position and the insertion position are stored.

It is determined whether or not the head combining process is necessary. If the insertion position is at the beginning of the entire chunk, the beginning join processing is unnecessary and control is passed to the processing.

Based on Meta-Q and Meta-A, the head join check is performed.

It is determined whether the combination is possible, and if the combination is not possible, the next process is skipped.

The inserted chunk and the preceding chunks are combined into one chunk.

It is determined whether the tail joining process is necessary. If the insertion position is at the end of the entire chunk, the tail join processing is not necessary, so control is transferred to the processing.

The tail combination check is performed based on Meta-Q and Meta-A.

It is determined whether the combination is possible, and if the combination is not possible, the next process is skipped.

Combine the inserted chunk and the chunk that follows it into a single chunk.

It is determined whether the intermediate joining process is necessary. If the extracted position is the beginning or end of the entire chunk, intermediate join processing is not necessary, so control is transferred to the processing.

Based on Meta-Q and Meta-A, an intermediate join check is performed.

It is determined whether the combination is possible, and if the combination is not possible, the next process is skipped.

The chunks before and after the extracted position are combined into one chunk.

It is checked whether the result of the above combining process is equal to the correct answer, and it is determined whether the final goal has been reached. If the final goal is not reached, transfer control to the process,
When the final goal is reached, control is passed to the processing.

The contents of the new Meta-Q are displayed on the display as the next problem, control is returned to the process, and the same process is repeated.

 The final result is displayed and the combining process ends.

 Next, a specific example of the combining process will be described with reference to FIG.

For example, the problem material (correct answer) is "AAAPQPQPQCX" as shown in Fig. 3 (a), and the problem that has been broken up is "PQ, A, X, C" as shown in Fig. 3 (c). , PQPQ, AA ”. These chunks are represented by Me in Figure 3 (b).
Like ta-A, the sequence is associated with the correct answer.

In this initial state, it is assumed that there is an instruction to insert "A" before "PQ" as shown in FIG. Meta-A and Meta-Q will be replaced by Meta-A1 and Me, respectively.
Copied as ta-Q1. Since the insertion position of "A" is the head, the head joining process is omitted.

In the tail joining process, the joining process of "A" and "PQ" is checked as shown in FIG. 3 (e). This check is performed as follows, for example.

In Meta-A1, in order to consider the combination that becomes “APQ”, first look at it from the left and focus on “A”. Next, search for "PQ" and if there is, bring it after "A" as shown in Fig. 3 (f).

The sequence of the entire Meta-A1 (Fig. 3 (g)) at this time is
Check if it is equal to the correct sequence. Since this case is the same, as shown in Figure 3 (h), "A" and "PQ"
And If they are not equal, go back and
On the contrary, paying attention to "PQ", in Meta-A1, "A"
Try adding "PQ" in front of it. This operation is repeated until the sequence of Meta-A1 is equal to the correct answer. If there is no case that is equal to the correct answer in the whole combination,
Judge that it cannot be combined.

In this example, by combining “A” and “PQ”, Meta-Q
1 becomes as shown in FIG. 3 (i).

In the next intermediate join process, the join between "APQ" and "X" in Meta-Q1 is the same as in the tail join process.
-Checked using A1. Here, it is determined that the combination is impossible.

As described above, the results after the combining process are as shown in FIGS. 3 (k), (l), and (m). Next, with this result as the initial state, the combining process is repeated in the same manner.

〔The invention's effect〕

As described above, according to the present invention, when associating individual chunks, only one solution is obtained without allowing a plurality of solutions, so that the amount of memory consumed can be reduced. The calculation time can be significantly reduced. Therefore, it can be applied to a small computer such as a so-called personal computer. When applied to a word game such as anagram, the user sometimes makes unexpected combinations, and can provide interesting learning material.

[Brief description of drawings]

FIG. 1 is a block diagram of the principle of the present invention, FIG. 2 is an explanatory view of the processing of one embodiment of the present invention, FIG. 3 is an example of a combining processing according to one embodiment of the present invention, and FIG. An explanatory diagram and FIG. 5 are explanatory diagrams of the correspondence relationship between the question material and the question. In the figure, 10 is a processing device, 11 is an input device, 12 is a display, 13
Is a problem material data storage unit, 14 is a problem chunk list storage unit, 15 is a chunk correspondence information storage unit, 16 is a problem presenting unit, 17 is a chunk move processing unit, 18 is a head join processing unit, 19 is a tail join process Reference numeral 20 is an intermediate joining processing unit, 21 is a goal determination unit, and 22 is a result display unit.

Claims (1)

(57) [Claims] 1. Regarding a problem in which a character string as a problem material is divided and rearranged by a lump of characters or a partial character string, an operation of extracting one of the lumps in the problem and inserting the lump in another position From the relationship with the problem material of, the arrangement order includes a central processing unit that performs processing to combine the adjacent masses partially corresponding to the problem material, and a device including a memory, including the inserted mass. A head join processing means (18) for checking the associativity between the lump and the lump in front of the insertion position and, if merging is possible, for merging those adjacent lumps into one lump. ， Checking the connectivity between the inserted lump containing the lump and the lump after the insertion position, and if it is possible to join, the adjacent lumps The tail joining processing means (19) that performs the process of joining the blocks into one block, and the position where the above blocks are removed is checked for the connectivity of the two blocks before and after, and if they can be combined, they are adjacent to each other. A rearranged partial character string characterized by comprising an intermediate joint processing means (20) for consolidating those chunks into one chunk and performing the joint processing only in the portion where the rearrangement has changed. Join processing method.