Patent Application
20220215681
This application claims priority to Japanese Patent Application No. 2021-001193 filed Jan. 7, 2021, the entire contents of which are incorporated herein by reference.
The disclosure relates to a technique for determining and evaluating a user's academic or other ability and the like using a computer.
Computer based testing (CBT) in which a test is conducted using a computer for determining and evaluating knowledge, ability, and the like of a user (examinee) has been widely used. In CBT, a question is displayed on a display of a terminal (computer). The user operates input devices, such as a mouse and a keyboard provided at the terminal, to answer a question displayed on the display.
JP 2011-81024A discloses a system for performing a writing test conducted using a CBT method with a tablet terminal that can receive handwriting input.
JP 2011-81024A is an example of background art.
However, when conducting a writing test using a CBT method, a handwriting input device is required for a user to input characters by handwriting.
Also, depending on the recognition accuracy of characters handwritten by a user, the correct answer may be determined as an incorrect answer, or conversely, an incorrect answer may be determined as the correct answer. When the correct answer is if (“Toyotomi Hideyoshi” in Chinese characters), for example, even if the characters handwritten by the user are correct, the character
(pronounced: “tomi”) may be recognized as the character
(pronounced: “kyo”), and may be determined to be an incorrect answer. In contrast, even if the characters handwritten by the user is
(transcription: “Toyokyo Hideyoshi”), which is an incorrect answer, the character
may be erroneously recognized as the character
, and determined to be the correct answer. If a character handwritten by a user is misrecognized, the user's knowledge, ability, and the like cannot be properly determined and evaluated.
In addition, when a writing test is conducted using a CBT method, if the writing test is conducted using a known conversion function from kana (Japanese kana syllabograms) to kanji (Chinese characters), the above-described problem does not occur. However, if the kana character string is (transcription: “Toyotomi Hideyoshi”), which was input by the user as the reading,
is output as a conversion candidate, but
is not output. Accordingly, even a user who incorrectly remembers the Chinese character notation of
as
will select
as an answer. Therefore, when a writing test is conducted using a CBT method, even if a known kana-to-kanji conversion function is used, determination and evaluation of user's knowledge, ability, and the like may not be performed as appropriate.
One or more embodiments may provide a technique capable of appropriately determining and evaluating knowledge, ability, and the like of a user who takes the test, when a writing test is conducted using a CBT method.
A character input device according to one or more embodiments may be configured as follows.
A first character string obtainment unit obtains a first character string, in response to an input character string that has been input. A similar character extraction unit extracts similar characters having similar shapes as characters in the first character string. A second character string generation unit generates one or more second character strings in which some or all of the characters in the first character string are replaced with similar characters extracted by the similar character extraction unit. Then, a conversion candidate output unit outputs the first character string and the second character strings as conversion candidates for the input character string.
In the above-described configuration, a first character string and a second character string are output as conversion candidates; the first character string is a character string whose reading is a kana character string (input character string) that is input by a user who takes a test of computer based testing (CBT) by operating a keyboard and the like, and the second character string is a character string obtained by replacing a part of characters in the first character string with characters similar in shape to the part of characters. The reading of the second character string may or may not be a kana character string input by a user. If the input character string that is input by the user is (transcription: “Toyotomi Hideyoshi”), for example, several conversion candidates are output, such as
which is the correct answer,
(transcription: “Toyokyo Hideyoshi”),
(transcription: “Toyotomi Yuyoshi”),
(transcription: “Toyotomi Hageyoshi”),
(transcription: “Toyokyo Hageyoshi”), and
(transcription: “Toyokyo Yuyoshi”). In the above-described example,
is the first character string, and,
,
, and
are the second character strings.
Accordingly, if the user incorrectly remembers the Chinese character notation of the input kana character string (reading), the user may select the correct conversion candidate (the first character string) by chance, but in most cases, the user will select an erroneous conversion candidate (the second character string). Therefore, in a case where the writing test is conducted using a CBT method, it is possible to appropriately determine and evaluate the knowledge, ability, and the like of a user who takes the test.
Furthermore, in the above-described configuration, because a user who takes the test of a CBT does not perform handwriting input, it is not necessary to provide a handwriting input device.
In addition, for example, the character input device may include an arrangement adjustment unit configured to adjust an arrangement order of the first character string and the second character strings that are output from the conversion candidate output unit as the conversion candidates for the input character string. With the above-described configuration, the position of the correct answer in the output conversion candidates may be changed for each question, and the user's knowledge, ability, and the like may be determined and evaluated more appropriately.
The first character string obtainment unit may be configured to obtain a predetermined character string as the first character string if the input character string is a specified character string, and to obtain a character string whose reading is the same as that of the input character string as the first character string if the input character string is not the specified character string.
When the input character string input by the user is not , which is the correct answer, but
(transcription: “Oda Nobunaga”), which is an incorrect answer, the first character string obtainment unit obtains
(“Oda Nobunaga” in Chinese characters) as the first character string. In the above-described case, as the second character strings,
(transcription: “Shokuda Nobunaga”),
(transcription: “Shokuda Kitsunaga”),
(transcription: “Oda Kitsunaga”),
(transcription: “Oda Kurumanaga”),
(transcription: “Shokuda Kurumanaga”), and the like are generated. Accordingly, in the above-described configuration, the user is not aware that the answer he/she has input, that is to say,
is an incorrect answer. As a result, the user's knowledge, ability, and the like may be determined and evaluated more appropriately.
The character input device may include a grading unit configured to grade, as an answer, a conversion candidate selected from the conversion candidates that have been output from the conversion candidate output unit.
According to one or more embodiments, in a case where a writing test is conducted using the computer based testing (CBT) method, it may be possible to appropriately determine and evaluate the knowledge, ability, and the like of a user who takes the test.
Hereinafter, one or more embodiments will be described.
The user terminal 2 is a personal computer or a tablet terminal operated by a user who takes a CBT writing test. The server device 1 distributes a question or questions of a writing test conducted using a CBT method to the user terminal 2, and grades an answer or answers transmitted from the user terminal 2.
In the present example, a character input device according to one or more embodiments is used as the server device 1.
The server device 1 distributes one or more questions of a writing test conducted using a CBT method to the user terminal 2.
The user terminal 2 displays the question or questions distributed from the server device 1 on a screen of a display of the user terminal 2. The user inputs an answer to the question or questions by operating a keyboard serving as an input device. In the present case, the user inputs an answer in the form of a kana (i.e., Japanese character) string.
The server device 1 distributes the following question to the user terminal 2, for example.
Whose personality is described in the following Japanese tanka (short poem of seventeen kana characters)?
(English translation: “If the small cuckoo does not sing for me, I will make the small cuckoo sing.”, a famous Japanese tanka in seventeen kana characters; “Na-ka-nu-na-ra, na-ka-se-te-mi-se-yo, ho-to-to-gi-su”)
A user who takes the test inputs a kana character string such as (transcription: “Toyotomi Hideyoshi”) or
(transcription: “Oda Nobunaga”) to the user terminal 2. The user terminal 2 outputs the input kana character string to the server device 1.
The server device 1 generates conversion candidates, according to the kana character string received from the user terminal 2. The server device 1 returns the generated conversion candidates to the user terminal 2. When the kana character string received from the user terminal 2 is “,” for example, the server device 1 generates conversion candidates such as “
,” “
,” “
,” “
,” “
,” and “
.” The correct answer to the question is the character string: “
.” The character string “
,” corresponds to a first character string according to one or more embodiments. The kana character strings:
,
,
, and
are conversion candidates obtained by replacing some characters of the correct answer
if with characters having similar shapes. The kana character strings:
,
, and
correspond to second character strings according to one or more embodiments.
The server device 1 may also create a conversion candidate in which all the characters of the correct answer are replaced with characters having similar shapes.
If the kana character string received from the user terminal 2 is (“Oda Nobunaga”), the server device 1 returns, to the user terminal 2, as conversion candidates,
,
,
, and the like. Although
is not the correct answer to the question,
is a conversion candidate whose reading is the character string input by the user. Accordingly,
corresponds to a first character string according to one or more embodiments.
The kana character strings: ,
, and the like are conversion candidates obtained by replacing some characters of
with characters having similar shapes. The kana character strings:
,
,
and the like correspond to the second character strings according to one or more embodiments.
The user terminal 2 displays the conversion candidates returned from the server device 1 on the screen of the display of the user terminal 2. The user operates a mouse or a keyboard serving as an input device to select his/her answer from the displayed conversion candidates. The user terminal 2 outputs, to the server device 1, the conversion candidate selected by the user as the answer of the user.
The server device 1 grades whether the user's answer to the question is the correct answer or an incorrect answer for each user terminal 2 (that is to say, for each user).
As described above, in the network system, if the user incorrectly remembers the Chinese character notation of the input kana character string (reading), the user may select the correct conversion candidate by chance, but in most cases, the user will select an erroneous conversion candidate. Therefore, when the writing test is conducted using a CBT method, it is possible to appropriately determine and evaluate the knowledge, ability, and the like of a user who takes the test.
In addition, in the above-described network system, because the user does not input an answer by handwriting, the user's knowledge, ability, and the like are not erroneously determined and evaluated depending on the recognition accuracy of the characters input by handwriting. Furthermore, the user does not need to prepare a tablet that can receive handwriting input.
The control unit 11 controls the units of the main body of the server device 1. In addition, the control unit 11 includes a first character string obtainment unit 11a, a similar character extraction unit 11b, a second character string generation unit 11c, an arrangement adjustment unit 11d, and a grading unit 11e. The first character string obtainment unit 11a, the similar character extraction unit 11b, the second character string generation unit 11c, the arrangement adjustment unit 11d, and the grading unit 11e that are included in the control unit 11 will be described later.
The similar character DB12 is a database in which characters having similar shapes are grouped and stored. The similar character DB12 is, for example, a database in which character groups (groups of characters having similar feature amounts of shapes) extracted as candidates are grouped and stored for a certain character recognized in character recognition by a known optical character reader (OCR). The dictionary DB13 is a database in which readings (kana character notation), parts of speech, converted character strings (Chinese character notation), and the like are registered in association with one another. The communication unit 14 performs data communication with the user terminals 2 connected via the network 5.
The similar character DB12 and the dictionary DB13 may be, for example, hard disk drives (HDDs), solid state drives (SSDs), or other storage media. Furthermore, the similar character DB12 and the dictionary DB13 may be configured by dividing a storage area in a single storage medium, or may also be configured by different storage media.
Next, the first character string obtainment unit 11a, the similar character extraction unit 11b, the second character string generation unit 11c, the arrangement adjustment unit 11d, and the grading unit 11e that are included in the control unit 11 will be described.
The first character string obtainment unit 11a obtains conversion candidates whose reading is the kana character string received from the user terminal 2. If the kana character string received from the user terminal 2 is the reading of the correct answer to the question, the first character string obtainment unit 11a obtains the Chinese character notation of the correct answer to the question as the first character string. If the kana character string received from the user terminal 2 is not the reading of the correct answer to the question, the first character string obtainment unit 11a obtains, as the first character strings, conversion candidates whose reading is the received kana character string.
For each of the characters belonging to the first character string obtained by the first character string obtainment unit 11a, the similar character extraction unit 11b extracts a character or characters similar in shape to that character. The similar character extraction unit 11b extracts characters having similar shapes by looking them up in the similar character DB12.
The second character string generation unit generates, as the second character string, a character string obtained by replacing some or all of the characters belonging to the first character string obtained by the first character string obtainment unit 11a with the characters extracted by the similar character extraction unit 11b. The second character string generation unit generates one or more second character strings.
The arrangement adjustment unit 11d adjusts the arrangement order of the first character string obtained by the first character string obtainment unit 11a and the second character strings generated by the second character string generation unit 11c when the first and second character strings are displayed as conversion candidates on the user terminal 2.
The grading unit 11e performs grading including correctness of the answer, based on the conversion candidate selected at the user terminal 2.
The control unit 11 of the server device 1 includes a hardware CPU, a memory, and other electronic circuits. When the hardware CPU executes the character input program according to one or more embodiments, the hardware CPU operates as the first character string obtainment unit 11a, the similar character extraction unit 11b, the second character string generation unit 11c, the arrangement adjustment unit 11d, and the grading unit 11e. The memory has an area for developing the character input program according to one or more embodiments, and an area for temporarily storing data and the like generated during execution of the character input program. The control unit 11 may also be an LSI in which the hardware CPU, the memory, and the like are integrated. The hardware CPU is a computer for executing the character input method according to one or more embodiments.
The control unit 21 controls the operations of the units of the main body of the user terminal 2.
The display 22 displays a screen corresponding to the state of the main body of the user terminal 2. The operation unit 23 includes input devices such as a mouse, a keyboard, and a touch panel. The operation unit 23 receives an input operation by the user to the main body of the user terminal 2. The communication unit 24 performs data communication with the server device 1 connected via the network 5.
The control unit 21 of the user terminal 2 includes a hardware CPU, a memory, and other electronic circuits. The control unit 21 may also be an LSI in which the hardware CPU, the memory, and the like are integrated.
Hereinafter, the operations of the server device 1 and the user terminal 2 when a writing test is conducted using a CBT method will be described.
The server device 1 distributes a question of a writing test from the communication unit 14 to the user terminal 2 (step s1). The question distributed from the server device 1 is received by the communication unit 24 of the user terminal 2 via the network 5. The server device 1 distributes the following question to the user terminal 2, for example.
Whose personality is described in the following Japanese tanka (short poem of seventeen kana characters)?
(English translation: “If the small cuckoo does not sing for me, I will make the small cuckoo sing.”, a famous Japanese tanka in seventeen kana characters; “Na-ka-nu-na-ra, na-ka-se-te-mi-se-yo, ho-to-to-gi-su”)
When the communication unit 24 of the user terminal 2 receives the question distributed from the server device 1, the user terminal 2 displays the received question on its display 22 (steps s21 and s22). The user operates the input device such as the keyboard that is provided in the operation unit 23 to input the answer to the question displayed on the display 22 in the form of a kana character string (reading). The user inputs, for example, ,
, or the like. In the present example,
may lead to the correct answer, and
will lead to an incorrect answer.
The user terminal 2 transmits the input reading (kana character string) from the communication unit 24 to the server device 1 (steps s23 and s24). The user terminal 2 waits for receiving, from the server device 1, conversion candidates (step s25).
After distributing the question to the user terminal 2 in the above-described step s1, the server device 1 waits to receive a reading from the user terminal 2 (step s2). When the communication unit 14 of the server device 1 receives the reading transmitted from the user terminal 2 in step s24, the server device 1 obtains a first character string (step s3). If the reading that has been received in step s3 is the reading of the correct answer to the question distributed in step s1, the server device 1 obtains the Chinese character notation of the correct answer as the first character string. In contrast, if the reading that has been received in step s3 is not the reading of the correct answer to the question distributed in step s1, the server device 1 obtains the Chinese character notation of the reading received as the first character string. When the received reading is the correct answer , for example, the server device 1 obtains
, which is the correct Chinese character notation of the correct answer as the first character string. In contrast, when the received reading is
, which is not the correct answer, the server device 1 obtains, as the first character string, any one of
and
(all of which can be read in Japanese as “Oda Nobunaga”). When the received reading is not the correct answer, the server device 1 obtains the first character string using the dictionary DB13. The first character string obtainment unit 11a performs the process related to step s3.
The server device 1 extracts a character having a similar shape, for each character belonging to the first character string obtained in step s3 (step s4). If the first character string obtained in step s3 is , for example, the server device 1 extracts Chinese characters having a similar shape to
, Chinese characters having a similar shape to
, Chinese characters having a similar shape to
, and Chinese characters having a similar shape to
. If the first character string obtained in step s3 is
, for example, the server device 1 extracts Chinese characters having a similar shape to
Chinese characters having a similar shape to
, Chinese characters having a similar shape to
, and Chinese characters having a similar shape to
. The server device 1 extracts the similar characters in step s4 using the similar character DB12. The similar character extraction unit 11b performs the process in step s4.
The server device 1 generates a predetermined number of second character strings in which some or all of the characters of the first character string obtained in step s3 are replaced with characters having similar shapes extracted in step s4 (step s5). When the first character string obtained in step s3 and the predetermined number of second character strings to be generated is five, the server device 1 generates, as the second character strings, e.g.
,
,
, and
. Also, when the first character string obtained in step s3 is
and the predetermined number of second character strings to be generated is five, the server device 1 generates, as second character strings, e.g.
,
and
The second character string generation unit 11c performs the process related to step s5.
The server device 1 sets the first character string obtained in step s3 and the predetermined number of second character strings generated in step s5 as conversion candidates for the answer to the question. When the user inputs an incorrect reading, the correct answer is not included in the conversion candidates of the answer to the question.
The server device 1 adjusts the arrangement order of the conversion candidates of the answer to the question (step s6). Here, the arrangement order is an order in which the conversion candidates are displayed on the display 22 of the user terminal 2. In step s6, the arrangement order of the conversion candidates is adjusted so that the location in which the correct answer is displayed randomly changes for each question. The arrangement adjustment unit 11d performs the process related to step s6.
The server device 1 transmits the conversion candidates of the answer to the question to the user terminal 2 that has transmitted the reading received by the server device 1 in step s2 (step s7). At the present time, the server device 1 instructs the user terminal 2 to display the conversion candidates of the answer to the question in the arrangement order adjusted in step s6.
When the communication unit 24 of the user terminal 2 receives the conversion candidates for the answer to the question, the user terminal 2 displays the received conversion candidates on its display 22 in the order instructed by the server device 1 (steps s25 and s26). When the user terminal 2 receives an operation of selecting one of the conversion candidates displayed on the display 22 through the operation unit 23, the user terminal 2 transmits the selected conversion candidate to the server device 1 as the answer to the question (steps s27 and s28), and then returns to the process in step s21.
Upon receiving the answer from the user terminal 2, the server device 1 grades the answer (steps s8 and s9), and then returns to the process in step s1.
Therefore, in the above-described network system, if the user incorrectly remembers the Chinese character notation of the reading that was input to the user terminal 2, then the user may select the correct conversion candidate by chance, but in most cases, the user will select an incorrect conversion candidate. Therefore, when the writing test is conducted using a CBT method, it is possible to appropriately determine and evaluate the knowledge, ability, and the like of the user who takes the test.
In addition, in the network system of the present example, the user does not input the answer by handwriting, thus the user's knowledge, ability, and the like are not erroneously determined and evaluated depending on the recognition accuracy of characters input by handwriting. Furthermore, the user does not need to prepare a tablet that can receive handwriting input.
In the above description, the server device 1 distributes questions one by one to the user terminal 2. However, the server device 1 may also be configured to distribute a plurality of questions at the same time to the user terminal 2, and to receive a reading and an answer for each question from the user terminal 2.
The control unit 21 of the user terminal 2 may also be provided with the configuration related to the first character string obtainment unit 11a, the similar character extraction unit 11b, the second character string generation unit 11c, the arrangement adjustment unit 11d, and the grading unit 11e that are included in the control unit 11 of the server device 1 described in the above example. With the above-described configuration, the user terminal can also be used as a learning terminal on which a user performs preliminary learning and review.
In the above example, the similar character extraction unit 11b is configured to extract, from existing Chinese characters, a Chinese character similar in shape to a Chinese character used in an answer to a question. However, for example, the similar character extraction unit 11b may also be configured to extract Chinese characters that are obtained by changing a part of a Chinese character used in the correct Chinese character notation as a character having a similar shape (the generated Chinese character may also be a non-existing Chinese character).
One or more embodiments may be applicable to a writing test of all types of test and a writing test of all subjects that are conducted using a CBT method.
One or more embodiments may not be limited to the above-described embodiment. One or more embodiments may be embodied by modifying the constituent elements without departing from the scope. Furthermore, various inventions may be formed by appropriately combining a plurality of constituent elements disclosed in the above embodiments. Some constituent elements may be deleted from all the constituent elements shown in one or more embodiments, for example. Furthermore, the constituent elements of different embodiments may also be combined as appropriate.
The correspondence relationship between the configuration according to one or more embodiments and the configuration according to the above-described embodiment may be described as the following supplementary note.
A character input device (1) including:
| Number | Date | Country | Kind |
|---|---|---|---|
| 2021-001193 | Jan 2021 | JP | national |
