JP2007226460A - Data processor and data processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To evaluate a document content based on a user's viewpoint. <P>SOLUTION: An evaluation reference setting part sets one or more evaluation references for evaluating the document content as "vocabularies" by input from the user. A term setting part sets one or more terms with respect to one vocabulary as "correspondent terms" by input from the user. A term sorting part extracts the term from an impression mail in which student's impression of a class is described, and determines whether or not the extracted term agree with any of the correspondent terms. A statistic part indexes content tendencies in an aggregate of impression mails as vocabulary values with each vocabulary as a reference. An evaluation display part makes a vocabulary distribution graph screen 290 to be displayed and the content tendencies in the aggregate of impression mails to be displayed in a graph. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for evaluating document contents.

In recent years, with the spread of computers and the advancement of network technology, the exchange of electronic information via the network has become popular. As a result, many of the business processes that have been conventionally performed on a paper basis are being replaced by network-based processes. Advances in digitalization and network technology have drastically reduced information acquisition costs. Under such circumstances, a technique for automatically analyzing the contents of a large amount of information is desired.
JP 2004-252632 A JP 2003-18324 A Tsukamoto Shinichi and Akahori Shinji, “Class Improvement by Collecting and Analyzing Learner Responses Using Mobile Phones”, Journal of Educational Systems Information Society, Japan, Educational System Information Society, 2004, 21, 3, p. 214-222

  As an example, a spam filter that determines the suitability of the contents of an electronic mail by natural language processing and automatically eliminates junk mail can be cited. The spam filter determines whether or not the e-mail is a junk e-mail by determining the appropriateness of each word included in the e-mail. However, such a spam filter merely determines whether or not an e-mail is appropriate in a binary manner. The inventor paid attention to the fact that when reading a sentence, the reader comprehensively evaluates the content of the document from various viewpoints set consciously and unconsciously.

  The present invention is an invention made based on the above-mentioned attention of the inventor, and its main object is to provide a technique for evaluating document contents based on a user's viewpoint with a simple user interface. It is in.

One embodiment of the present invention is a data processing apparatus.
First, this apparatus sets one or more evaluation criteria and one or more corresponding terms for each of the evaluation criteria by input from a user. Then, it is determined whether the term in the document received from the external device is a corresponding term set for any evaluation criterion, and the appearance frequency of the corresponding term is determined for each evaluation criterion for the set of document data to be evaluated. calculate.

  By such processing, the content tendency in the document data set can be indexed by the appearance frequency of the corresponding term based on the evaluation criteria set by the user. The “term” used herein may be a word, a word group constituting a phrase, or a byte stream.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

  According to the present invention, document contents can be analyzed from a viewpoint set by a user with a simple user interface.

  In this embodiment, a lesson evaluation system is taken as an example. First, an outline of the lesson evaluation system is described, and then “vocabulary analysis processing” which is a main feature of the present invention is described.

FIG. 1 is a hardware configuration diagram of a lesson evaluation system.
The lesson evaluation system 10 is a system for a student to evaluate lesson contents, assuming introduction to an educational institution such as a university. The class evaluation system 10 is a system used by teachers and students, and provides a mechanism for feeding back student responses to classes to class contents.

  The class evaluation system 10 includes a class evaluation apparatus 100, a mail server 202, and a plurality of client terminals 204. These are connected to each other via the Internet 200. The lesson evaluation apparatus 100 is an apparatus used by a teacher (hereinafter, “teacher” is also referred to as “user” in the sense of clearly indicating that it is a user of the lesson evaluation apparatus 100). Each client terminal 204 is a general communication terminal equipped with a mail transmission / reception function, such as a mobile phone or a notebook computer used by students.

  First, the teacher sets an e-mail address dedicated to the course when the course is opened (hereinafter referred to as “course address”). For example, a course address such as “fraA@xxx.co.jp” is set for a course consisting of 12 classes “French A”. A student who takes a class in French A sends and receives an e-mail to and from the class evaluation apparatus 100 via the course address. For example, when a student attends a class, an “attendance mail” for notifying attendance is transmitted to the course address. The mail server 202 transfers the attendance mail received from the client terminal 204 to the lesson evaluation apparatus 100. The mail server 202 may be a general known device. The class evaluation apparatus 100 automatically counts which students attended which course by attendance mail.

In addition to this, the student transmits an impression email describing the impression of the class to the course address. The class evaluation apparatus 100 performs natural language analysis on the impression mail and quantitatively evaluates the degree of understanding and motivation of the students. For example, if the impression mail contains many pre-defined “motivated words” such as “done”, “done”, and “want to do”, it can be determined that the student has a high level of understanding and willingness to learn. Thus, the lesson evaluation apparatus 100 shown in the present embodiment quantifies the response of the student to the lesson by analyzing the terms of the impression mail. Such processing is referred to as “vocabulary analysis processing” in this specification.
Various information obtained from student e-mails is used as useful information for improving class content.

FIG. 2 is a functional block diagram of the lesson evaluation apparatus.
Each block shown here can be realized in hardware by an element such as a CPU of a computer or a mechanical device, and in software it is realized by a computer program or the like. Draw functional blocks. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by a combination of hardware and software.

The class evaluation device 100 includes a user interface processing unit 110, a communication unit 120, a data processing unit 130, and a data holding unit 150.
The user interface processing unit 110 is in charge of processing related to the entire user interface, such as input processing from a teacher user and information display for the user. The communication unit 120 is in charge of communication processing with the client terminal 204 via the mail server 202. The data processing unit 130 executes various types of data processing based on data acquired from the user interface processing unit 110 and the communication unit 120. The data processing unit 130 also serves as an interface among the user interface processing unit 110, the communication unit 120, and the data holding unit 150.
The data holding unit 150 stores various data such as setting data prepared in advance and an e-mail received from the data processing unit 130. The communication unit 120 includes a mail transmission / reception unit 122 that processes transmission / reception of mail.

  The user interface processing unit 110 includes a display unit 112 and an input unit 118. The input unit 118 receives an input operation from the user. The display unit 112 displays various information to the user. Display unit 112 further includes an evaluation display unit 114. The evaluation display unit 114 displays various user interface screens for vocabulary analysis processing in the display format shown in FIG.

The data holding unit 150 includes a mail data holding unit 152, a student data holding unit 154, a lesson data holding unit 156, and an evaluation reference data holding unit 158.
The mail data holding unit 152 holds an e-mail received from a student. The student data holding unit 154 holds student data. The student data is data registered in advance for the student, such as the department, age, gender, and student number. The lesson data holding unit 156 holds lesson data. The lesson data is data related to lessons that are aggregated based on information obtained from e-mails such as student attendance, test results, and impressions of lessons for each course. The evaluation reference data holding unit 158 holds evaluation reference data. The evaluation reference data is data in which the correspondence relationship between “vocabulary” and “corresponding term” is defined in advance, and this will be described in detail in the explanation related to FIG. 7 later.

The data processing unit 130 includes an email processing unit 132 and a lesson data management unit 138.
The mail processing unit 132 analyzes the contents of the electronic mail. The mail processing unit 132 includes a classification unit 134 and a matching determination unit 136. The classification unit 134 performs classification according to the contents of the email. As already mentioned, e-mails received from students are various such as attendance e-mails and impression e-mails. The classification unit 134 classifies the electronic mail according to the content. For example, an e-mail in which the word “attendance” is included in “subject” is classified according to a predetermined classification condition such as a condition that it is an attendance mail.

  The matching determination unit 136 determines whether the received electronic mail is valid mail as a processing target. For example, it can be said that an attendance email received before the start of a lesson is not valid. The matching determination unit 136 determines the validity of the received electronic mail according to a predetermined matching condition. Invalid email is excluded from analysis.

  The lesson data management unit 138 manages data related to courses as lesson data for each lesson. For example, various data such as an average score of an examination, a student's understanding level analyzed from an impression mail, and a lesson plan are managed as lesson data. The lesson data management unit 138 includes a student management unit 140, a setting unit 142, and an analysis unit 164. The student management unit 140 manages the student's learning state such as attendance rate and test results. The setting unit 142 sets various conditions related to the analysis of the e-mail such as the conditions of the matching determination unit 136 and the classification unit 134.

  The setting unit 142 further includes an evaluation criterion setting unit 144, a term setting unit 146, and a document attribute setting unit 148. By the vocabulary analysis process, the content tendency of the comment mail group received from the student is indexed. The evaluation criterion setting unit 144 sets the evaluation criterion as “vocabulary”. For example, when it is desired to index the degree of understanding, motivation, satisfaction, etc. of students, four vocabularies of “knowledge”, “interest”, “achievement”, and “insight” may be set as evaluation criteria. “Knowledge” here is the name of an evaluation standard for analyzing impression mail from the viewpoint of “whether the student understands the content of the lesson”. The vocabulary setting will be described in more detail in connection with FIG.

  The term setting unit 146 sets terms belonging to each vocabulary (hereinafter referred to as “corresponding terms”). For example, when the impression mail contains many terms “I knew”, “I learned”, and “I learned”, it can be estimated that the student has steadily acquired “knowledge” from the class. In this case, these terms are set as the corresponding terms of “knowledge” for the vocabulary “knowledge” set earlier. Depending on what terms are included in the impression mail, the content tendency based on each vocabulary is indexed as a “vocabulary value” described later. In summary, the content trend of the impression mail group is indexed from the relationship between the terms included in the impression mail, the corresponding terms, and the vocabulary. A specific processing process will be described in detail with reference to FIG.

  The document attribute setting unit 148 sets the document attribute for classifying the impression mail according to the input by the user. For example, if a collection of feedback emails from male students and a collection of feedback emails from female students are subject to lexical analysis processing separately, whether the sender of the response emails is a male student or a female student is a document attribute It becomes.

  The analysis unit 164 performs vocabulary analysis processing on the impression mail group according to various setting conditions by the setting unit 142. The analysis unit 164 includes a term classification unit 160 and a statistics unit 162. The term classification unit 160 acquires the impression mail from the mail processing unit 132 and extracts the term from the comment mail. Term extraction may be performed by applying a technique such as known morphological analysis. Next, the term classification unit 160 determines whether the extracted term matches any of the corresponding terms set by the term setting unit 146. Here, the term “match” may be a complete match or a concept including partial matches such as forward match or backward match. The statistical unit 162 counts the number of appearances of the corresponding term for the set of impression emails to be evaluated. Then, the appearance frequency of the corresponding term is calculated as a “vocabulary value” for each vocabulary. In the previous example, the vocabulary value was indexed to which vocabulary the terms included in the impression email were biased among the four vocabularies of “knowledge”, “interest”, “achievement”, and “insight”. Although it is a numerical value, the details of the calculation method will be described later. The vocabulary values are displayed in a graph in the format shown in FIG.

FIG. 3 is a screen view of the menu screen of the lesson evaluation apparatus.
When a user who is a teacher logs in to the lesson evaluation apparatus 100, first, the menu screen 210 is displayed. The course selection area 212 displays a list of currently opened courses. Here, three courses of “French A2005”, “French B2005” and “French C2005” are opened. In the figure, “French A2005” is selected.

  When one of the buttons included in the menu button group 214 is selected, a screen for accessing various lesson data about the course selected in the course selection area 212 is displayed. The attendance book button 216 is a button for displaying the attendance status of students. The impression acceptance button 218 is a button for displaying a list of the contents of the impression mail received from the student. The submission management button 220 is a button for displaying the submission status such as homework. The test result button 222 is a button for displaying a list of test results. The consistency check button 224 is a button for displaying a screen for setting consistency conditions. The medical record button 226 is a button for displaying information on the student's lesson attitudes and grades as “karte”. The notification button 228 is a button for displaying a screen for notifying notification items related to lessons such as absence notice. The voting button 230 is a button for displaying a screen for collecting students' opinions.

  Here, in order to explain only the basic functions of the lesson evaluation apparatus 100, the menu is displayed when the attendance book button 216, the impression reception button 218, and the test result button 222 are selected from the button group of the menu button group 214. The attendance book screen 240, the impression reception screen 250, and the test result screen 260 will be described. Then, the vocabulary analysis process will be mainly described.

FIG. 4 is a screen view of the attendance book screen.
The attendance book screen 240 is displayed when the attendance book button 216 is clicked on the menu screen 210 of FIG. The teacher shows the secret words of the class to the students at the start of the class. The secret word may be shown verbally in the classroom. The student who attended the class inputs this password and sends an attendance email to the course address. The classification unit 134 classifies the attendance mail, and the matching determination unit 136 determines whether the correct password is described in the attendance mail. If it is a valid attendance email, the lesson data management unit 138 recognizes the student who is the sender of the attendance email as a regular attendee. The display unit 112 displays the names of students attending the class on the attendance list screen 240.

  The secret word is a mechanism to prevent students who are not actually attending from attending. As a variation for attendance management, an attendance management method based on the position of the client terminal 204 is also conceivable. For example, when the client terminal 204 is equipped with a GPS (Global Positioning System) function, the attendance mail may include position information of the client terminal 204 at the time of transmission. Based on this position information, the alignment determination unit 136 determines whether it is an attendance mail from a student who is actually in the classroom. In addition to GPS, the client terminal 204 may transmit the connected base station ID as location information.

The class selection area 242 displays the attendance status regarding each class of the “French A2005” class. According to the figure, the lesson of this course is the fourth, and the attendance rate for the four lessons is listed. In the lesson selection area 242, the fourth lesson of the theme “pronunciation of R” is selected. The attendance status display area 244 displays a list of names of students attending the fourth class selected in the class selection area 242. Here, of the 6 students registered in the course, 5 students except for “Takeharu Yoshiharu” attend the class.
When the lesson evaluation apparatus 100 accepts the attendance mail, the display unit 112 automatically updates the contents of the attendance book screen 240, so that the teacher can grasp the attendance status of each student in real time and accurately using the attendance book screen 240.

FIG. 5 is a screen view of the test result screen.
The test result screen 260 is displayed when the test result button 222 is clicked on the menu screen 210 of FIG. When the test is performed during the class, the user inputs the scoring result via the user interface processing unit 110. The student may send an answer email describing the answer to the exam to the course address. Moreover, the lesson evaluation apparatus 100 may be provided with a scoring function. The scoring results are displayed on the test result screen 260 as the results of each student's test.

  The test selection area 262 displays the result of each test related to the “French A2005” course. According to the figure, the past five tests have been carried out. In the test selection area 262, the fourth test result on June 22 is selected. The test result list area 264 displays a list of student test results for the fourth test selected in the test selection area 262. Test results include scores, rankings, changes in scores compared to the previous test, and the like.

  The test results can also be saved as a CSV (Comma Separated Values) format file. For example, in the case of the student “Hideyuki Ishida”, data such as “80 (point)” and “2 (rank)” is set as an attribute value for each item such as “score” and “rank”. The lesson data management unit 138 generates a test result file in CSV format from student data and lesson data according to a predetermined format.

FIG. 6 is a screen view of an impression acceptance screen.
The impression acceptance screen 250 is displayed when the impression acceptance button 218 is clicked on the menu screen 210 of FIG. When the class is over, the student sends an impression email to the course address. This feedback email allows the teacher to immediately know the student's reaction to the lesson. The lesson selection area 252 displays the reception status of the impression mail regarding each lesson of the “French A2005” course. In the class selection area 252, the fourth class is selected. The impression list area 256 displays a list of student's impressions regarding the fourth class selected in the class selection area 252. Furthermore, the impression display area 254 displays the impression of the student “Mayumi Arai” selected in the impression list area 256.

  Here, when the teacher sets the keyword “question”, the student whose word “question” is included in the impression mail is marked with an asterisk. This feature allows teachers to easily find feedback emails that contain questions.

  Here, the impression emails of three students are displayed in a list, but when there are dozens of students or the number of letters in the impression email is large, it is not easy to grasp the reaction of each student to the lesson. . The vocabulary analysis function of the lesson evaluation apparatus 100 shown in the present embodiment can index the content tendency in a set of comment emails and visually display this as a graph.

FIG. 7 is a screen diagram of the vocabulary registration screen.
In order to index the content trend in the set of impression mails, first, the evaluation display unit 114 displays the vocabulary registration screen 270 shown in FIG. The vocabulary set input area 272 is an area for designating a vocabulary set name. Here, a vocabulary set named “Class Evaluation” is set in order to find out how students respond to class content. The vocabulary name input area 274 sets one or more vocabularies for the vocabulary set in the vocabulary set input area 272. Here, four vocabularies of “knowledge”, “insight”, “interest”, and “achievement” are set. When a term included in the impression mail is included in any of the semantic categories of these four vocabularies, the term is subjected to lexical analysis processing. Here, the content of the impression email is based on four perspectives: whether the student has knowledge, demonstrates insight into the content of the lesson, is interested in the lesson, or feels attainment It is assumed that such a vocabulary is set in order to index the tendency. The corresponding term input area 276 is an area for setting a corresponding term for the vocabulary selected in the vocabulary name input area 274.

  In the figure, corresponding terms such as “I knew” and “I knew” are set for the vocabulary “knowledge”. This indicates that terms such as “know” and “remembered” are included in the concept of the vocabulary “knowledge”. That is, when terms such as “I knew” or “I learned” are extracted from the impression mail, the statistical unit 162 adds a predetermined value, for example, “1” to the vocabulary value for the vocabulary “knowledge”. For example, when the terms “know” and “remembered” are detected 10 times and 30 times, respectively, from the set of comment emails, the vocabulary value is calculated as “40” by 10 + 30. Thereby, the vocabulary value is calculated for each vocabulary by the addition method based on the terms included in the comment mail group, the set vocabulary and the corresponding terms. By using the vocabulary value, the content trend of the impression mail group can be indexed from the evaluation standard called vocabulary. The calculation method of the vocabulary value is not limited to this, and any numerical value may be used as long as the content tendency is indexed based on each vocabulary in the impression mail group to be evaluated. For example, the number of appearances of corresponding terms / the number of impression mails, the number of appearances of corresponding terms / the total number of characters in the impression mails may be calculated as vocabulary values.

  A difference in contribution to the vocabulary value may be provided even between corresponding terms belonging to the same vocabulary. For example, the term “know” may be more relevant to the vocabulary “knowledge” than the term “helpful”. In that case, a score of 2 points may be set for the term “know” and 1 point may be set for the term “reference”. Then, when the term “know” is extracted from the impression mail, “2” is added to the vocabulary value of “knowledge”, and when the term “reference” is extracted, “1” is added. May be. According to such a processing method, it is possible to perform vocabulary analysis processing that reflects even the nuances of terms.

  On the vocabulary registration screen 270, vocabularies and corresponding terms defined in the evaluation standard data held by the evaluation standard data holding unit 158 may be set. In this evaluation standard data, a vocabulary “knowledge” and corresponding terms “know” and “remembered” may be defined in advance. In this case, when the user selects the vocabulary “knowledge” from among the vocabularies defined in the evaluation reference data, the corresponding terms “know” and “learned” are automatically entered in the corresponding term input area 276. Is set. The user may newly add a corresponding term to the existing “knowledge” vocabulary, or may exclude an existing corresponding term automatically set in the corresponding term input area 276 from being set. If the evaluation reference data is prepared in advance, the input burden on the user's vocabulary registration screen 270 can be reduced. In particular, the convenience of the vocabulary analysis function can be improved for users who are not familiar with the operation of the lesson evaluation apparatus 100.

  By the way, if the term “recognized” is included in the impression email, this term is not set as the corresponding term of “knowledge”, and this term is reflected in the vocabulary value of “knowledge”. There is no. However, it can be said that one of the corresponding terms “know” and the extracted term “recognized” are in a synonym relationship. The term classification unit 160 may hold synonym dictionary data in which synonym relationships between terms are established. Then, the term classification unit 160 refers to the synonym dictionary data, and if the term extracted from the impression mail does not match the corresponding term but corresponds to the synonym of the corresponding term, it is regarded as the corresponding term belonging to the vocabulary. Also good. In this example, the term “recognized” is a synonym of “know”, so that the lexical value of “knowledge” is added. According to such a processing method, more terms than can be set in the corresponding term input area 276 can be included in the evaluation target, so that more precise vocabulary analysis can be performed. In addition, it is effective in reducing the user input burden on the vocabulary registration screen 270.

FIG. 8 is a screen diagram of the attribute-specific aggregation setting screen.
Vocabulary analysis processing may be performed on all of the feedback emails received from students, but for example, a set of feedback emails to be evaluated is identified, such as boys and girls, high graders and low graders There are cases where you want to. Through the attribute-specific tabulation setting screen 280, the user can specify an impression mail to be evaluated.

  When the attribute-specific check area 282 is checked, impression emails are classified. The classification standard is called “document attribute”. An attribute name input area 284 is an area for setting document attributes. Here, the gender of the sender is set as the document attribute. That is, the vocabulary analysis processing is executed for each of the comment mail group from the boy student and the comment mail group from the girl student. In addition to this, the document attributes can also be viewed from various viewpoints such as “higher grades and lower grades”, “review emails of 100 characters or more, impression emails of 50 to 99 characters, impression emails of 49 characters or less”. It can be set.

The display attribute setting area 286 is an area for selecting a document attribute to be displayed on the vocabulary distribution graph screen 290 shown in FIG. 9 based on the set document attribute. Here, the vocabulary analysis processing result for each of the male student's comment mail group and the female student's comment mail group is set to be displayed on one screen.
After the setting on the vocabulary registration screen 270 and the setting on the attribute-specific tabulation setting screen 280 are made, the analysis unit 164 receives the impression mail from the mail processing unit 132 and starts the vocabulary analysis processing.

FIG. 9 is a screen diagram of the vocabulary distribution graph screen.
In the vocabulary distribution graph screen 290, as set in the attribute-specific tabulation setting screen 280 of FIG. 8, the result of each vocabulary analysis process is divided into a group of male student feedback mails and a girl student feedback mail. It is displayed. The vocabulary distribution graph screen 290 uses the four vocabularies set on the vocabulary registration screen 270 of FIG. 7 as “knowledge”, “insight”, “interest”, and “achievement” as evaluation axes, and the vocabulary value for each vocabulary is displayed. A graph is displayed as a parameter. According to the figure, boys are better at learning, achievement, and insight than girls, but girls are more interested in classes.
According to such a vocabulary analysis and display method, it is possible to index and visualize a student's response to a lesson from an arbitrary viewpoint from a free format impression mail.

FIG. 10 is a screen view of the impression analysis screen.
This impression analysis screen 292 is displayed when a predetermined item is selected from the menu of the impression reception screen 250 of FIG. The vocabulary set selection area 294 on the impression analysis screen 292 displays a list of vocabulary sets set on the vocabulary registration screen 270. Here, two types of vocabulary sets of “class evaluation” and “sensory expression” are shown. The vocabulary name list area 296 shows details of the vocabulary set “class evaluation” selected in the vocabulary set selection area 294. In this vocabulary set “class evaluation”, vocabulary values for four vocabularies of “knowledge”, “interest”, “achievement”, and “insight” are shown. Further, the vocabulary distribution graph screen 290 in the same figure is as described in relation to FIG.

FIG. 11 is a flowchart showing the basic processing steps of vocabulary analysis.
Here, a description will be given assuming that the settings on the vocabulary registration screen 270 and the attribute-specific tabulation setting screen 280 have been completed. First, the mail transmission / reception unit 122 receives an impression mail from the student, and the mail processing unit 132 transfers an appropriate impression mail to the term classification unit 160. In this way, the term classification unit 160 acquires the impression mail (S10). If the document attribute is set on the attribute-specific tabulation setting screen 280, the term classification unit 160 classifies the impression mail according to the document attribute (S12).

  The term classification unit 160 extracts terms from the impression mail (S14). When this term corresponds to the corresponding term (Y in S16), the statistical unit 162 updates the vocabulary value of the vocabulary to which the corresponding term belongs (S20). Even when it is not a corresponding term (N in S16), if it is a synonym of the corresponding term (Y in S18), the vocabulary value of the vocabulary to which the corresponding term belongs is updated (S20). When it is neither a corresponding term nor a synonym (N at S16, N at S18), the process at S20 is skipped.

  When a term is further included in the impression mail acquired in S10 (Y in S22), the process returns to S14, and the term is extracted again, and the same processing is repeated. On the other hand, when a new term to be extracted is not included in the comment mail (N in S22), the term classification unit 160 requests the mail processing unit 132 for a comment mail to be processed next (S24). If there is an impression mail to be processed next (Y in S24), the process returns to S10 to acquire the next impression mail, and the same process is repeated. If there is no comment mail to be processed next (N in S24), the evaluation display unit 114 displays the vocabulary distribution graph screen 290 using the vocabulary value for each vocabulary as a parameter (S26).

The present invention has been described above based on the embodiments.
According to the lesson evaluation apparatus 100 shown in the present embodiment, a student's response to a lesson can be quantitatively evaluated from an impression mail obtained when the lesson evaluation system 10 is operated. In fact, students with lower grades are mostly “interesting”, and students with higher grades are often “knowledge”. Therefore, the vocabulary analysis processing as described above makes it easy to find out the actual understanding level of the students. Inexperienced teachers tend to neglect the check of students' responses because they place more emphasis on class progress. According to the lesson evaluation apparatus 100 shown in the present embodiment, the student's reaction to the lesson can be grasped immediately, objectively, and visually, so the lesson content can be flexibly changed according to the student's reaction. Information about the degree of understanding and willingness to class, the personality of students, etc. are useful for various decisions such as classification, class level adjustment, and necessity of supplementary classes.

  As the document attribute, the impression mail may be classified based on the time, such as the impression mail for the first lesson and the impression mail for the second lesson. According to such a classification method, the number of lessons can be repeated and how the response of the students to the lesson can be grasped over time. This is also useful information for teacher personnel evaluation. For example, when the student's motivation is improved every time the lesson is repeated, such a fact can be indexed, so that it is easy to appropriately evaluate a teacher who conducts a class that improves the student's motivation.

  As the results of the operation of the class evaluation system 10 are accumulated, there is a possibility that various knowledge about the relationship between student responses, vocabulary, and corresponding terms may be obtained. For example, typical terms that are often used when students can't keep up with the class may be discovered. If such terms are registered as evaluation standard data in advance, even a teacher who uses the class evaluation apparatus 100 for the first time uses the experiences of various teachers in the past to find students to be treated early. It becomes easy to do. Therefore, there is a possibility that the evaluation standard data itself brings commercial value.

  With the recent declining birthrate, it is certain that the number of universities that will break the capacity will increase in the future. In order to survive in such a situation, the demand for quality of lessons has never been higher. Even in view of such a social background, the lesson evaluation apparatus 100 shown in the present embodiment is effective as a tool for improving the quality of lessons.

  In the present embodiment, the vocabulary analysis function has been described using the lesson evaluation system 10 as a subject. However, it goes without saying that the vocabulary analysis function described in the present embodiment is not limited to the lesson evaluation system 10 and can be widely applied. . For example, an application of quantitative analysis of the individuality of each sentence author by setting vocabulary such as “confirmation” or “ambiguous” in a court decision sentence or business document can be considered. As described above, the present invention can be widely applied to a scene where the contents of a document are evaluated. Note that the characteristic lexical analysis function in this embodiment may be provided as a plug-in for existing document processing software.

The function of the document data acquisition unit described in the claims is mainly realized by the mail transmission / reception unit 122 in this embodiment.
It should be understood by those skilled in the art that the functions to be fulfilled by the constituent elements described in the claims are realized by a single function block or a combination of the functional blocks shown in the present embodiment.

  The present invention has been described based on the embodiments. This embodiment is an exemplification, and it will be understood by those skilled in the art that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are also within the scope of the present invention. is there.

For example, the following modifications can be considered.
1. In this embodiment, the vocabulary distribution graph is displayed as a four-axis graph corresponding to the four vocabularies of “knowledge”, “achievement”, “insight”, and “interest”. However, the vocabulary distribution graph need not be limited to four axes, and may have more axes. For example, the vocabulary distribution graph may be formed as a five-axis graph such as “knowledge”, “interest”, “experience”, “hearing”, “consultation”. The vocabulary “insight” may be further classified into two types of vocabulary, “foresight” and “reflection”. In such a case, a new axis may be set not only according to “insight” that is the upper vocabulary but also according to “foresight” and “reflection” that are lower vocabularies.

  2. Although FIG. 9 shows the vocabulary distribution for the entire electronic mail acquired in each class, the vocabulary distribution graph screen 290 may display the vocabulary distribution per class. By comparing the vocabulary distribution for each class, it is possible to visually confirm how the vocabulary distribution of the class changes as the number of classes progresses.

  3. In the vocabulary distribution graph screen 290 of FIG. 9, the value of each axis is set in the range of 0-100. However, the range of this axis may be appropriately adjusted according to the size of the vocabulary distribution graph. For example, in the case of FIG. 9, since “interest” has a value of “80” for a girl student and “60” for a boy student, the axis value of “interest” is set in the range of 0-80. Also good. According to such display control, the graph can be displayed in an appropriate size regardless of the size of the vocabulary value. Moreover, the vocabulary value may be normalized so that it may fall in the range of 0-100. Even with such a processing method, the graph can be displayed in an appropriate size regardless of the size of the vocabulary value.

  4). On the vocabulary distribution graph screen 290, the appearance frequency of the corresponding term for each vocabulary in the electronic mail group of the entire class is displayed as a vocabulary value. On the other hand, the student position for each vocabulary may be displayed in a graph. For example, the vocabulary values of “knowledge”, “achievement”, “insight”, and “interest” are calculated from the comment mail group of the student A, and the centroid of the vocabulary distribution graph formed from these four vocabulary values is calculated. The position of the student A may be specified. Then, the position distribution of each student may be displayed on the vocabulary distribution graph screen 290. According to such a processing method, the term usage tendency for each student can be checked together on one screen. For example, it is possible to check how many students have reached the level of insight, not just “interest”, and who they are.

  5. If the impression email contains a sentence pattern that denies the previous statement such as "I thought it was but I knew it was not", the former "thought" reflects the current student's feelings It can be said that it is not a written word. In such a case, the term classification unit 160 may exclude “thinking” from the vocabulary value calculation based on sentence dependency. In this way, it is possible to calculate the vocabulary value according to the actual situation by calculating the vocabulary value in consideration of the relationship rather than simply calculating the vocabulary value from the terms included in the impression mail. .

  6). On the vocabulary distribution graph screen 290, for example, when an area near the “insight” axis is clicked, among the students registered in the course, the names of students whose corresponding terms of “insight” appear frequently are displayed. For example, the top five students who frequently use the corresponding term “insight” may be displayed in a list. When one of the displayed student names is clicked, medical chart data indicating the learning status of the student is displayed.

  7). The relationship between the vocabulary and the corresponding term shown in the vocabulary registration screen 270 of FIG. 7 may be filed. Then, different setting files may be used according to students, teachers, or lesson contents. For example, terms that appear frequently in the feedback emails of students with poor grades and those that appear frequently in feedback emails of students with good grades may show different tendencies. In such a case, “a term that appears frequently only among students with good grades” may be assigned as a corresponding term for “knowledge” or “insight”. Similarly, “terms that appear frequently only among students with poor grades” may be assigned as the corresponding terms of “interest” and “achievement”. As a result, as lessons are accumulated, it is expected that appropriate relationships between vocabulary and corresponding terms will be established. Moreover, it is thought that a frequent term changes also according to the kind of course. For example, the term used in the impression email may be different for a science subject such as mathematics or physics and a liberal subject such as ancient or art history. Based on these points, the relationship between vocabulary and corresponding terms may be set according to the subject.

  8). Even if the student A and the student B use the same term “I understand” in the impression mail, the degree of understanding of the two is not necessarily the same. For example, “I understand” of student A who has excellent grades has a high possibility of actually understanding the contents of the class, but “I understand” of student B who has poor grades may only have a superficial understanding. unknown. In such a case, “I understand” of student A may be set as a corresponding term of “insight”, and “I understand” of student B may be set as a corresponding term of “knowledge”. Alternatively, Student A's “Okay” adds one point of “insight” vocabulary value, but Student B's “Okay” may add “insight” vocabulary value of 0.5 points. Thus, the reliability of information obtained from students may be set according to the students. This reliability may change in real time according to the student's class attitude and test results.

  9. The class evaluation device 100 may have a consultation function. For example, a student whose vocabulary changes from “insight” to “interest” each time the number of classes overlaps may be unable to keep up with the content of the class. A student whose number of comment letters is increasing and whose vocabulary of “interest” is particularly increased may have a high level of motivation for class but may not be fully satisfied. The consultation function of the lesson evaluation apparatus 100 displays a warning or advice message to the teacher when a predetermined condition is satisfied for a change in vocabulary distribution. The conditions for executing such a consultation function and the content of the message may be set in a file (hereinafter referred to as “consultation file”). Even inexperienced teachers can use a consultation file set by experienced teachers to conduct classes while receiving support from the experience of other teachers.

  10. Depending on the results of the personality test, students may be classified into several types. Depending on the type of student, the terminology used and the appropriate teaching methods will vary. For example, an A type student and a B type student may easily exhibit a synergistic effect of enhancing each other's learning effect, but an A type and a C type may be counterproductive. In addition, there may be teachers who are good at type A, and teachers who are good at type B. The class evaluation apparatus 100 may classify students into several types according to the words used by the students, and may consider the results in classification, selection of a teacher in charge, and the like. In addition, depending on the type of students and the degree of understanding, a consultation may be made regarding what kind of tasks should be given at what timing.

  11. You may compare not only students taking the same course at the same time, but also students taking this year's course with those taking last year's course. For example, if the change in the vocabulary distribution of student A this year is equivalent to the change in the vocabulary distribution of student B last year, the future change and grade of the vocabulary distribution of student A will be determined to some extent depending on the data of student B. It may be predictable. When there are several students corresponding to the student B in the past, an improvement in prediction accuracy can be expected. If student A's grade improves more than expected, the teacher's teaching ability should be highly appreciated. In this way, comparing students with the same tendency is effective in evaluating the teacher's teaching ability.

It is a hardware block diagram of a lesson evaluation system. It is a functional block diagram of a lesson evaluation apparatus. It is a screen figure of the menu screen of a lesson evaluation apparatus. It is a screen figure of an attendance book screen. It is a screen figure of a test result screen. It is a screen figure of an impression reception screen. It is a screen figure of a vocabulary registration screen. It is a screen figure of an attribute total setting screen. It is a screen figure of a vocabulary distribution graph screen. It is a screen figure of an impression analysis screen. It is a flowchart which shows the basic process of a vocabulary analysis.

Explanation of symbols

  10 class evaluation system, 100 class evaluation device, 110 user interface processing unit, 112 display unit, 114 evaluation display unit, 118 input unit, 120 communication unit, 122 mail transmission / reception unit, 130 data processing unit, 132 mail processing unit, 134 classification , 136 Consistency determination unit, 138 Class data management unit, 140 Student management unit, 142 Setting unit, 144 Evaluation criteria setting unit, 146 Term setting unit, 148 Document attribute setting unit, 150 Data holding unit, 152 Mail data holding unit, 154 Student data holding unit, 156 Class data holding unit, 158 Evaluation standard data holding unit, 160 Term classification unit, 162 Statistics unit, 164 Analysis unit, 200 Internet, 202 Mail server, 204 Client terminal, 210 Menu screen, 240 attendance screen, 250 impression reception screen, 260 test result screen, 270 vocabulary registration screen, 280 attribute-specific tabulation setting screen, 290 vocabulary distribution graph screen, 292 impression analysis screen.

Claims (8)

An evaluation criterion setting unit for setting one or more evaluation criteria for evaluating the document content by input from the user;
A term setting unit that sets one or more terms as corresponding terms by input from a user for one evaluation criterion;
A document data acquisition unit for receiving document data from an external device;
A term classifying unit that extracts terms from the received document data and determines whether the extracted terms are corresponding terms set for any of the evaluation criteria;
For a set of document data to be evaluated, a statistical unit that calculates the appearance frequency of corresponding terms for each evaluation criterion,
A data processing apparatus comprising:   An evaluation display unit that visually displays a content trend in the set of document data to be evaluated by a graphic formed using the appearance frequency of the corresponding term for each evaluation criterion as a parameter. The data processing apparatus according to 1. A document attribute setting unit for setting a document attribute as a document data classification criterion by an input from a user;
The statistical unit classifies the received document data according to document attributes, and calculates the appearance frequency of corresponding terms for each evaluation criterion for each set of document data classified according to each document attribute. The data processing apparatus according to claim 1 or 2.   The term classification unit refers to predetermined synonym dictionary data that defines a synonym relationship between terms when the extracted term does not correspond to a corresponding term of any evaluation criterion, and the extracted term is any evaluation criterion 4. The data processing apparatus according to claim 1, wherein the extracted term is regarded as a corresponding term for the evaluation criterion when the corresponding term is a synonym for the corresponding term. The document data acquisition unit receives, as document data, an e-mail that describes a student's impressions about the class,
5. The data processing according to claim 1, wherein the evaluation criterion setting unit sets one or more evaluation criteria as an evaluation criterion for indexing a content tendency of an impression on a student's class. apparatus. A plurality of types of evaluation criteria, and an evaluation criteria data holding unit for holding predetermined evaluation criteria data in which one or more corresponding terms are defined for each of the evaluation criteria,
The evaluation standard setting unit selects, as a setting target, an evaluation standard specified by a user from among a plurality of types of evaluation standards defined in the evaluation standard data,
The data processing according to any one of claims 1 to 5, wherein the term setting unit sets one or more corresponding terms defined for the selected evaluation criterion as corresponding terms for the evaluation criterion. apparatus. Setting one or more evaluation criteria for evaluating document contents by input from a user;
Setting one or more terms as a corresponding term by input from a user for one evaluation criterion;
Receiving document data from an external device;
Extracting terms from the received document data and determining whether the extracted terms are corresponding terms set for any evaluation criteria;
Calculating a frequency of appearance of corresponding terms for each evaluation criterion for a set of document data to be evaluated;
A data processing method comprising: A function for setting one or more evaluation criteria for evaluating the document content by input from the user;
A function for setting one or more terms as a corresponding term by input from a user for one evaluation criterion;
A function of extracting a term of document data received from an external device and determining whether the extracted term is a corresponding term set for any of the evaluation criteria;
A function for calculating the appearance frequency of corresponding terms for each evaluation criterion for a set of document data to be evaluated,
A data processing program characterized by causing a computer to exhibit.

Images