JP2005173061A - Exercise assistance system and exercise assistance method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve a problem in which it is difficult for a teacher to give hints and explanations beforehand for all problems although high educational effects are obtained by implementing many problems in an education method like case study. <P>SOLUTION: In an exercise wherein a plurality of learners try to solve one problem, a place in the problem to which a fixed number of learners give annotations is detected and the place is shown as a hint to learners who do not give annotations to the place. A place to which only a small number of learners give annotations is detected and judged as a place which learners less understand or a place which is misunderstood, and the place is presented to the teacher. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to the field of computer-aided learning support systems. More specifically, the present invention relates to a learning support system in which a plurality of learners can learn while using terminals connected by a computer, and teachers can also participate in the support of learners. And an exercise support method.

  In human resource development in companies, it is necessary not only to acquire knowledge through textbooks and lectures, but also to ensure that learners have a deep understanding until they can practice knowledge. Case studies are often used as an educational method for acquiring such practical skills. One method of case study is to use multimedia data such as documents, videos, still images, etc., assuming work situations, and to extract points that learners have noticed from these multimedia data Is well done. In addition to extracting the noticed points, the scene that is the extraction target may be identified, or a keyword indicating from what point of view may be associated. In other words, it is an operation of extracting information according to the theme of learning. The extracted work itself uses the learned knowledge, so it is an exercise that helps to establish knowledge.

  In such exercises, it is important that learners themselves first work. On the other hand, if the learner does not have the learned concept properly, information cannot be extracted well and the willingness to learn may be reduced. In addition, when referring to the learning results of other learners, the process of thinking for themselves disappears and the learning effect is diminished. Therefore, a method for presenting an appropriate hint has been proposed in order to eliminate the learning deadlock.

  In the learning support method described in Patent Literature 1, when the learner presses a help button or the like, the learning support system detects a deadlock in learning, searches the teaching material database for explanations accumulated in advance, and learns. It can be shown to the person.

  In addition, according to the hint display method in the learning system described in Patent Document 2, a method is provided in which hints can be given to the learner in stages. That is, in a learning system that creates an answer to a question sentence using a diagram, a teacher prepares hints for each task in advance for each stage. When the learner presses the “hint display” button, the figure of the question sentence and the figure of the hint prepared in advance are exchanged. At this time, since a hint is prepared for each stage, the level of the hint to be displayed is changed each time the button is pressed. By inserting hints into the question sentence, the learner can focus on the task while referring to the appropriate level of hints.

  Patent Document 3 discloses a data clustering method.

JP 2001-356678 A

JP 2000-98864 A JP-A-5-205058

  For issues such as case studies, it is effective to implement a number of issues. However, it is difficult for teachers to give appropriate tips and detailed explanations for all cases. Also, for learners, if they are unable to point out well and get stuck, their willingness to learn decreases. It may be possible to recognize that there is a different perspective just by showing the learner the correct answer or the learning outcome of another learner at an early stage, but refer to the answer before thinking about it. As a result, the learner cannot obtain a deep understanding of the task.

  Moreover, in the conventional technology, it is necessary for the teacher to prepare information for explanation and hints in advance.

  A first object of the present invention is to provide a learning support system capable of automatically or semi-automatically generating hint information that enhances a learner's willingness to learn without burdening the teacher when providing learning materials. It is to provide a practice support method.

  Another object of the present invention is to provide a learning support system and an exercise support method that can automatically determine the degree of understanding of a learner and can be used for advice given to the learner by a teacher.

  The present invention is an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network, and the exercise support system stores material data for learning as multimedia data A storage unit, a material data providing unit that outputs the material data to the learner's terminal in accordance with a request from the user terminal, and the material data displayed on the learner's terminal are given by the user An annotation information storage unit that stores annotation information, an annotation information totaling unit that classifies and aggregates the annotation information according to the annotation grant status, and a summary result of the annotation information including the grant status of other users A hint information generation unit that generates hint information for each learner.

  Further, the present invention is an exercise support method using an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network, and upon receiving a material information acquisition request from a user, Search for the corresponding material from the material database, output and display it on the user terminal, accept input of annotations associated with locations in the material data, store the input annotation information in the annotation information database, and allow multiple users Hint information is generated using a result of the annotation information including the status of the assignment, and the generated hint information is displayed on the learner's terminal that requested the hint information output.

  ADVANTAGE OF THE INVENTION According to this invention, in providing the learning material, the hint information which raises a learner's willingness to learn can be generated automatically or semi-automatically without imposing a burden on a teacher. For example, there is an exercise form in which each learner designates a place in the shared multimedia information and places a “mark” in accordance with the assignment presented by the teacher. According to the present invention, in such an exercise form, there is provided means or a method for tabulating according to the proximity of the place pointed out by the learner and displaying it as a hint to the learner who did not point out the place. . Thereby, since the learner can obtain an accurate hint corresponding to the degree of understanding, the motivation for learning increases. In addition, it is possible for the teacher to save the trouble of creating hints in advance.

  In addition, according to the present invention, it is possible to provide the teacher with information that can be used for advice by presenting the teacher with a place where the learner has little understanding or a possibility of being mistaken. For example, for a place where the number of learners pointed out is less than a certain number, it is determined that the place is difficult to find or wrong, and means for presenting the place to the teacher is provided. For this reason, since the points to be explained can be narrowed down and presented to the teacher, the teacher can efficiently give advice to the learner.

Hereinafter, embodiments of the invention will be described with reference to the drawings.
FIG. 1 is a diagram showing an overall outline of a usage pattern of a learning support system according to an embodiment of the present invention. This learning support system includes an exercise support server 103, a plurality of learner terminals 101, a teacher terminal 102, and the Internet and other networks 106 connecting these learner terminals, teacher terminals, and exercise support servers. . A material database 104, an annotation information database 105, and a client database 107 are connected to the exercise support server 103. Note that the learner terminal and the teacher terminal are collectively referred to as a user terminal.

  In this learning support system, a plurality of learners having respective ID codes connect to the exercise support server 103 via the Internet or other network 106 using the learner's terminal 101, and the exercise support server is a material database. The material data held in 104 is referred to on the learner's terminal. At this time, according to the assignment given by the teacher, “marks” are added to the material. Furthermore, the “viewpoint”, which is a keyword indicating from what point of view the mark is given, is specified in association with the mark. For example, in an exercise called “Danger Prediction Training at the Work Site”, illustrations depicting scenes of factories and office work are used as materials, and points that seem to be dangerous are pointed out in the scenes indicated by the materials. Practice methods are taken. In addition, a keyword indicating from what viewpoint the risk arises is also associated. By such work by the learner, the position of the mark given to the material and the viewpoint are associated with each other as “annotation information”, which is accumulated in the annotation information database 105 of the exercise support server. In the client database 107, information related to the learning status of each learner, information related to teacher advice, and the like are recorded.

  On the other hand, when the learner is stuck or wants a hint, the learner requests the practice support server 103 for a hint via the learner's terminal 101. In response to a request from the learner, the exercise support server aggregates the annotation information from a plurality of learners accumulated so far by the method described below, and the learner who requested the hint stumbles from any viewpoint. And where it is not recognized in the material, the stumble viewpoint or the unrecognized location is presented to the learner as hint information. For this reason, since it becomes possible to automatically generate hint information suitable for each learner, the teacher does not need to create hint information suitable for the subject of the exercise in advance.

  Further, when the exercise support server 103 receives a request for advice information display from the teacher terminal 102 used by the teacher, the viewpoint that only a small number of learners can be extracted from the accumulated annotation information by a method described later. Detects and places. By using this information, the teacher can recognize the viewpoint that the learner lacks understanding without referring to all the materials used in the assignment, and can efficiently refer to the places that are difficult to extract. become able to.

FIG. 2 shows a configuration diagram of a learning support system according to an embodiment of the present invention. FIG. 3 shows a processing flow of the exercise support server of FIG.
The learner's terminal 101 of this exercise support system includes a material data display unit 201, an annotation information input unit 202, and a hint information display unit 203. The exercise support server 103 includes a shared material data storage unit 204, an annotation information storage unit 205, an annotation information totaling unit 206, a hint information generation unit 207, an advice auxiliary information generation unit 208, and an exercise management unit 210. In addition, the teacher terminal 102 includes an input / output function unit including an advice auxiliary information display unit 209.

  In this example, there is provided a means for displaying advice information to the teacher. However, only the exercise support server and the learner terminal may be configured, and a plurality of teacher terminals are connected. It is good also as a structure. Moreover, it is good also as a structure provided only with the method of outputting advice information to the terminal for teachers, without producing | generating the hint information to a learner. In addition, the user of the learner terminal may be used by a teacher who inputs annotation information that is the original data of the hint information in advance or a user who assists learning.

  The flow of processing of the exercise support server 103 will be described with reference to FIG. When the exercise management unit 210 receives a request from a client (302), the exercise management unit 210 reads information corresponding to the client from the client database 107 (305). Then, the process performed by the exercise support server 103 is assigned according to the type of request from the learner's terminal and the teacher's terminal (306).

  If the learner requests display of the material via the learner's terminal, the shared material data storage unit 204 searches the material database 104 for the requested material data. The retrieved material data is transmitted to the learner's terminal and displayed on the material data display unit 201 (307).

  The learner uses the annotation information input unit 202 of the learner's terminal to specify the location and set the annotation for the displayed material. In the following description, the “viewpoint” input simultaneously at this time is described as being associated with the location where the annotation is set. However, it is not necessary to set the “viewpoint”. In the case where the viewpoint is not set, an equivalent effect can be obtained by treating it as a specific viewpoint set in the following description. An annotation setting method differs depending on whether the material is a video, a still image, a document file, or the like, which will be described later.

  The annotation information set by the annotation information input unit 202 is transmitted to the exercise support server 103 via the network 106 and received by the annotation information storage unit 205 via the exercise management unit 210. The received annotation information is stored in the annotation information database 105 (308).

  When a hint is requested, the annotation information totaling unit 206 totals the annotation information stored in the annotation information database 105 (309). A method of counting will be described later. The timing of counting is totaled at the timing when the annotation information is received in the annotation information storage unit, the timing when hint information is requested by the learner, the timing when advice auxiliary information is requested from the teacher, or at a fixed interval.

  From the learner's terminal 101, the learner requests to display a hint, the time after the start of learning exceeds a certain value, or the difference from the annotation given by another learner has become more than a certain value At the timing, hint information is displayed on the learner's terminal (311). At this time, the hint information generation unit 207 generates hint information corresponding to the target learner based on the result obtained by the annotation information totaling unit 206. The hint information display unit 203 presents the generated hint information to the learner by displaying it on the screen. A method for generating hint information will be described later. Here, two types of hints are output. In other words, the hint is related to the “point of view” that the learner has not successfully extracted and the “location” that the learner has not been able to extract well. These give the learner step-by-step hints by first generating hints for the first viewpoint, and then outputting hints for the second place once hints have been issued for all viewpoints. Can do.

On the other hand, the teacher is required to display advice auxiliary information from the teacher terminal, the time after the start of learning exceeds a certain level, or the number of annotations given by the learner is constant among learners The auxiliary advice information is aggregated (310) at the timing when the above difference occurs, and displayed on the teacher terminal. The advice auxiliary information generation unit 208 generates information for assisting the advice by a method described later based on the results obtained by the annotation information totaling unit 206 and displays the information on the advice auxiliary information display unit 209 (312).
The results and the like collected by the annotation information totaling unit 206 of the exercise support server 103 are recorded in the annotation information storage unit 205 and also recorded in the client database 107 or updated as client information for each learner or teacher. (313).

  FIG. 4 is an example of a screen in which when a video is used as a material, a learner refers to the material on the learner's terminal 101, adds an annotation, and requests and displays hint information. For example, the task given by the teacher is “listing the dangers lurking during the work from the video of the work site where the luggage is moved”. The learner stops the video where he / she thinks it is dangerous, specifies what viewpoint he / she thinks dangerous, and sets an annotation. When the learner wants a hint and clicks the “display hint” button, the hint information generated by the hint information generation unit 207 of the learning support server 103 is displayed on the screen.

  FIG. 5 is an example of annotation information held in the annotation information database when the material is video. The annotation given by the learner is assigned with an annotation ID for identifying the annotation. Furthermore, the location to which the annotation is added, the ID of the learner to which the annotation is added, and the viewpoint at the time of extraction are stored in association with each other. In the example of FIG. 5, the playback position of the video material is stored as the pointed-out point where the annotation is set. That is, the annotation A1 indicates that data indicating that the learner with the ID L1 has given the viewpoint of “working posture” is stored at the position 0 minute 58 seconds from the start of the video.

  Note that the table holding annotation information may be held for each material, or may be held as one table to which a field for storing material identifiers is added.

  FIG. 6 is an example of a screen in which when a still image is used as a material, the learner refers to the material on the learner's terminal, adds an annotation, and requests and displays hint information. The operation of designating a specific location in the material provides a screen that is performed by designating specific coordinates in the still image.

  FIG. 7 is an example of annotation information held in the annotation information database when a still image is used as a material. The pointed out part is stored as coordinates on the still image.

  When the material is a document, the annotation position is specified by combining the number of pages, rows, columns, and the like.

  The annotation information totaling unit 206 in FIG. 2 totals the annotation information collected as shown in FIGS. The annotation information totaling unit performs clustering so as to classify annotations pointing out the same part. There are various data clustering methods. For example, there is a clustering method according to Patent Document 3.

  Here, an example of a clustering method will be described with reference to FIGS. FIG. 8 shows a clustering method based on the distance between two points as an example of an algorithm for performing clustering. FIG. 9 is a schematic diagram showing the relationship between video data and annotations attached to the video data.

  FIG. 8 shows a method of combining annotations set closer to a predetermined threshold value as one “pointed point”. First, learner information is read (step 700), and a counter for designating the class and annotation currently being focused is initialized (step 701). A class Ui for collecting annotations is created (step 702), and the annotation Ai that is currently focused on is added to the class (step 703). If there is a next annotation (step 704), the distance between the currently focused annotation and the next annotation is measured. If this distance is smaller than the threshold value T1 determined separately, the next annotation is set as the annotation currently focused on (step 709).

  On the other hand, if the distance between the two annotations is larger than the threshold value T1 in step 705, the midpoint of the set Ui of pointed locations up to that point is held as the location specified by the set of pointed locations (step 706). ), Proceed to the next step (step 707). If there is no next annotation in step 704, the learner information is recorded and updated in the annotation information storage unit 205 and the client database 107, and the process ends. If there is a next annotation, the process returns to step 701.

  Note that the annotation that is the basis of the threshold is automatically generated based on the learner's operation. In other words, it is expected that the number of annotations will be small and the error will be large in the early stage when the number of learning is small. Therefore, an initial value of annotation information may be given based on a result obtained by preliminary operation of a teacher or other person who has a predetermined knowledge or more.

  The steps so far (steps 702 to 709) will be described in detail with reference to FIG. 9, using the annotation data shown in FIG. Assuming that the threshold T1 is 5 seconds, each of annotation IDs A1 to A4 has an interval of 1 to 4 seconds. Also, the position of the middle point “1 minute 1 second” of this class A1 to A4 is held as the position designated by this class.

  Similarly, A5 to A8 and A9 are grouped into the same class. Then, as shown in FIG. 9, annotations designated at close positions are grouped into one class.

  By such steps, basic data of the annotation information totaling table 801 as shown in FIG. 10 is created.

  Further, in step 708, a “viewpoint extraction ratio” representing how much each learner has extracted each viewpoint is calculated. This calculates the percentage of learners with viewpoint Vj in each pointed set Ui. In other words, if the number of learners is LN, and the number of viewpoints Ui given the viewpoint Vj is N (Ui, Vj), the extraction ratio of the viewpoint Vj is calculated by N (Ui, Vj) ÷ LN To do. This calculation result is entered in the annotation information totaling table 801 in FIG.

  FIG. 11 is an example of a flow of processing for giving a hint regarding the viewpoint to the learner Ln using the annotation information totaling table 801 of FIG. This process is a process in which a learner Ln who has requested a hint displays a viewpoint having a lower extraction ratio than other learners as a hint. First, learner information is read (step 900), and a counter indicating the viewpoint of interest is initialized (step 901). Next, the expected number C (Vj) of viewpoint Vj is counted (step 902). C (Vj) is the number of locations in the annotation information tabulation table 801 that are larger than the separately specified threshold value T2. That is, it is determined that a part extracted by a certain number of learners from the same viewpoint is a correct answer, and that viewpoint Vj should be extracted. For example, in the case of the annotation information table shown in FIG. 10, if the threshold value T2 = 0.4, the expected number of extractions of the viewpoint “working posture” is 2, and the material used in this task includes the viewpoint “working”. “Attitude” indicates that it is included in two places.

  A ratio Q (Ln, Vj) of the number N (Ln, Vj) of Vj extracted by the learner Ln with respect to the expected extraction number C (Vj) is obtained (step 903). When this ratio is smaller than the threshold value T3 set separately (step 904), a hint is displayed to the learner Ln (step 905). For example, when the viewpoint Vj is “work posture”, the hint displayed at this time is “Let's focus on the work posture”. If there are other viewpoints (step 906), the viewpoint for determining whether to display the hint is changed to the following (step 907), and the above steps are repeatedly executed.

  The hint to be displayed is generated with reference to the information of the previous learning record. If the learner learns for the first time, there is no information on the previous learning record, so the hint is presented only when the learner requests a hint. If there is no next viewpoint in step 906, the learner information is recorded and updated in the annotation information storage unit 205 and the client database 107, and the process ends. If the learner wishes to continue learning, the process returns to step 901.

  FIG. 12 is an example of a flow of processing for giving a hint related to a place to the learner Ln using the annotation information totaling table 801 of FIG. This process is a process for a learner who has requested a hint to detect a location that has not been successfully extracted compared to other learners and to display the location as a hint. First, learner information is read (step 900), and a counter indicating a place of interest and a viewpoint is initialized (steps 1001 and 1002). Next, when the extraction ratio of the viewpoint Vj exceeds the threshold value T4 specified separately for the indicated point Ui, the process proceeds to step 1004 (step 1003). Step 1003 is a process for determining whether or not the indicated point Ui is a point to be extracted as the viewpoint Vj. In order not to display as a hint, it is assumed that only a small number of learners have been extracted, or that the learner has made a mistake, or that it is a rare case that requires additional explanation. It is.

  Next, it is determined whether the learner Ln who has requested the hint has already extracted the viewpoint Vj (step 1004). If it cannot be extracted, this location Ui is a candidate for displaying a hint. Next, it is determined whether or not the indicated point Ui is extracted from a viewpoint other than the viewpoint Vj that is currently focused on (step 1005). If it has been extracted, a message that prompts the user to observe the vicinity of the indicated point Ui from another viewpoint is output (step 1006). On the other hand, if the viewpoint Vj is not extracted in step 1005, a message that prompts the user to refer to the pointed portion Ui with priority is output. Here, steps 1005 and 1006 may not be executed, but only step 1007 may be executed.

  Further, the hint location Ui is displayed on the screen referred to by the learner Ln (step 1008). The above steps are executed for all viewpoints and all locations (steps 1009 to 1012). If there is no next viewpoint in step 1011, the learner information is recorded and updated in the annotation information storage unit 205 and the client database 107, and the process ends. If the learner wishes to continue learning, the process returns to step 1001.

  When the material is a video, according to the method as described above, the hint information can be automatically displayed while the video is being reproduced. That is, in FIG. 10, when a certain range before and after each class is being played back, a message such as “focus on this place” is displayed on the screen of the learner's terminal to alert the learner. It becomes possible.

  It is also possible to skip and reproduce a certain range including a part to which a correct answer is already given, or to start reproduction from a part where a hint should be displayed next. By using such means, the already learned part is not referred to and the learning time can be shortened, so that the learner's willingness to learn can be maintained.

  FIG. 13 is a process for generating information to be provided to the teacher by using the annotation information totaling table 801 in FIG. 10 to collect information about which viewpoint the learner has not successfully extracted.

  First, teacher information is read (step 1100), and a counter indicating the point of interest is initialized (step 1101). Next, similarly to step 902, the expected number C (Vj) of viewpoints Vj is obtained (step 1102). The number of statements B (Vj) from which the learner has extracted Vj is counted at the points detected when counting the expected number of extractions. B (Vj) is obtained as follows. That is, taking the annotation information tabulation table 801 of FIG. 10 as an example, the estimated number C (Vj) of viewpoint “working posture” is two points U1 and U4 when the threshold value T2 = 0.4. It becomes. Further, the number of statements B (Vj) from which the learner has extracted the viewpoint “working posture” is 6, which is the number of points given the viewpoint “working posture” at the indicated points U1 and U4.

  Next, the extraction rate R (Vj) of the overall viewpoint Vj is obtained. When the number of learners is LN, this is obtained as the ratio of B (Vj) to the total number C (Vj) × LN that all learners should originally extract the viewpoint Vj (step 1104).

  If R (Vj) calculated here is less than the separately designated threshold value T5, it is determined that the viewpoint is difficult for the learner to extract, and is presented to the teacher (step 1106). The above steps are repeated for all viewpoints (steps 1107 and 1108). If there is no next viewpoint in step 1007, the teacher information is recorded and updated in the annotation information storage unit 205 and the client database 107, and the process ends. If the learner wishes to continue learning, the process returns to step 1101.

  FIG. 14 is a process for generating information to be provided to the teacher by using the annotation information totaling table 801 of FIG. 10 to totalize information about which part the learner has not successfully extracted.

  First, teacher information is read (step 1200), and a counter indicating a point of interest and a viewpoint is initialized (step 1201). Next, in the indicated point Ui, it is determined whether or not the extraction ratio of the viewpoint Vj is larger than 0.0 and smaller than a separately designated threshold value T6 (step 1203). By this step, it is possible to detect a place where one or more learners have extracted and only a certain number of learners or less have been extracted. The indicated point Ui detected in step 1203 is presented to the teacher together with the viewpoint Vj that is currently focused. The above steps are executed for all viewpoints and all locations (steps 1205-1208). If there is no next viewpoint in step 1207, the teacher information is recorded and updated in the annotation information storage unit 205 and the client database 107, and the process ends. If the learner wishes to continue learning, the process returns to step 1201.

  In the above example, when an annotation is designated, a certain point in the material is designated. In other words, a specific frame is specified when the material is an image, a specific coordinate is specified when the material is an image, and a specific page, the number of rows / columns is specified when the material is a document, etc. .

  On the other hand, when pointing out, it is also possible to specify with a certain predetermined range. When the material is a video, it is a start point and an end point, and when it is an image, it is an area surrounded by an arbitrary shape. In the case of a video, it is also possible to specify such as a frame to be pointed out and a region on the screen, a start point / end point of the pointed point and coordinates on the screen, a start point / end point and a region on the screen.

  By clustering the points to be pointed out, it is possible to generate hints for learners and advice information for teachers as in the above example. The clustering method will be described with reference to FIGS.

  In FIG. 15, an image is used as a material, and a start point and an end point are designated as indicated points in a predetermined range. In other combinations, the calculation can be similarly performed by expanding the one-dimensional axis in FIGS. 15 and 16 to two dimensions and three dimensions.

  First, a membership function representing the “probability” of the indication is defined with respect to the indicated location, with the indication range as the center as shown in FIG. Here, in order to simplify the explanation, it is assumed that the slope of the hypotenuse of the membership function is constant.

  A similar membership function is defined for each indicated point. When summing up the points indicated, first, the membership values are added together. The added result is, for example, as shown in FIG. In FIG. 16, there are four vertices from P1 to P4. Considering that a point above a certain threshold is “many people pointed out”, the ranges are set as classes C1, C2, and C3, respectively.

  When displaying the hint information, it is determined whether the learner who has requested the hint points out a location included in this class. If not, the range is output as a hint. The hint display method is presented to the learner by displaying a message “focus on this area” when entering this range during playback or by starting playback from the first point of each class.

  As advice assistance information for the teacher, by presenting a place (P4) that does not satisfy the threshold value among the vertices, assistance is provided for providing more appropriate advice.

Next, another embodiment of the present invention will be described with reference to FIGS.
FIG. 17 shows the flow of processing of the exercise support server 103 in this embodiment. In this embodiment, the presentation method of the next material database is changed based on the learning result of the previous learner. For example, in the embodiment described above, when an annotation is given as shown in FIGS. 5 and 9, it is determined that each learner understands well the time zone where the annotation ID is from A1 to A3. The This determination can be realized by the following algorithm, for example.

  That is, as a result of aggregation by the method described in FIG. 8 and FIG. 10, the “perspective extraction ratio” that indicates how much each learner has extracted each viewpoint is pointed out from the viewpoint corresponding to that part. This is used as a criterion for how much certainty there should be. That is, a search is made for a value above a predetermined threshold (for example, a value above 0.3). The point of view U1 corresponds to the viewpoint of “working posture”, the point of point U2 “not ready” and “work environment”, and the point of point U3 “part handling”.

  When the learner who has requested the output of the material data is L2, the indicated location U1 has already given the place where the “working posture” should be indicated. Therefore, presentation of material data in a predetermined range including the pointed portion U1 is omitted.

  In addition, regarding the remaining area, it may be possible to present an appropriate hint at an early stage. The point to be pointed out U2 should be pointed out from two viewpoints of “under preparation” and “work environment”, but the learner L2 points out only “under preparation”. Accordingly, in the vicinity of U2, a hint that there is another point of view, specifically, a hint that "let's review again from the viewpoint of the work environment" is displayed. The point to be pointed U3 should be pointed out from the viewpoint of “part handling”, but the learner L2 has not pointed out. Therefore, a hint about a place “Are you not aware of anything near here?” Is displayed.

  In this way, as shown in FIG. 17, the material database is re-edited in accordance with the learning level and understanding level of each learner, in other words, the frame area above the threshold is removed (1701). And stored in the client database 107. According to this method, the learner reflects the results of his / her learning and performs learning based on material data different from the previous one. Is thought to be enhanced.

It is explanatory drawing showing the utilization model of the exercise assistance system of this invention. It is a lineblock diagram of the practice support system of one embodiment of the present invention. It is a flow of processing of the exercise support server of one embodiment of the present invention. It is an example of the screen of the terminal for learners when the material used for the exercise is a video. This table holds annotation information when the material used for the exercise is a video. It is an example of the screen of the terminal for learners when the material used for the exercise is a still image. This table holds annotation information when the material used for the exercise is a still image. It is a figure which shows the flow which totals annotation information. It is a schematic diagram of the relationship between video data and annotations attached to the video data. It is a table after totaling annotation information. This is a process flow for outputting a hint related to the viewpoint to the learner. This is a process flow for outputting a hint about a place to a learner. It is a flow of processing for outputting advice auxiliary information for each extracted portion to the teacher. It is a flow of processing for outputting advice auxiliary information for each extracted portion to the teacher. It is a schematic diagram of how to hold a value when a range is specified as a method for specifying an extraction location. It is a schematic diagram of the clustering method when a range is designated as the extraction location designation method. It is a flow of processing of the exercise support server of another embodiment of the present invention. FIG. 18 is a schematic diagram of the relationship between video data and annotations attached to the video data in the example of FIG. 17.

Explanation of symbols

101 ... Learner's terminal 102 ... Teacher's terminal 103 ... Exercise support server 104 ... Material database 105 ... Annotation information database 106 ... Network 107 ... Client database 201 ... Material data display unit 202 ... Annotation information input unit 203 ... Hint information display unit 204 ... Shared material data storage unit 205 ... Annotation information storage unit 206 ... Annotation information totaling unit 207 ... Hint information generating unit 208 ... Advice auxiliary information generating unit 209 ... Advice auxiliary information display unit 801 ... Annotation information totaling table.

Claims (12)

An exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
The exercise support system includes a material data storage unit that stores multimedia data for learning as material data, a material data providing unit that outputs the material data to a learner terminal according to a request from the user terminal, An annotation information storage unit for storing annotation information given by a user in the material data displayed on the learner's terminal, an annotation information totaling unit for classifying and adding the annotation information according to an annotation grant status, and others And a hint information generation unit that generates hint information for each learner using the totaled result of the annotation information including the granting status of the plurality of users. An exercise support system in which a plurality of learner terminals and at least one teacher terminal are connected to an exercise support server via a network,
The exercise support system includes a material data storage unit that stores multimedia data for learning as material data, and a material data providing unit that outputs the material data to the learner terminal according to a request from the learner terminal. An annotation information storage unit that stores annotation information provided by a user in the material data displayed on the learner terminal; an annotation information totaling unit that classifies and adds the annotation information according to an annotation application state; An advice auxiliary information generating unit that detects a part where the number of users who have annotated from a total result of the annotation information given by a plurality of users is less than a certain number and generates the corresponding part as advice auxiliary information; An exercise support system characterized by An exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
The exercise support system includes an exercise management unit that sorts requests from learner terminals according to request types, a material database that stores multimedia data displayed on the learner terminals, and inputs from the user to the learner terminals. A material data storage unit that outputs material data according to the requested request, an annotation information input unit for the user to assign annotations associated with locations in the material data displayed on the learner's terminal, and the provided annotations An annotation information storage unit for receiving information by the exercise support server, an annotation information database for storing the assigned annotation information, an annotation information totaling unit for classifying the stored annotation information according to the location information to which the annotation is provided, The result of annotation information In addition, a hint information generation unit that detects a part where a certain number of users have annotated and the user who requested the hint has not been granted and outputs the part as a hint, and learning the generated hint information An exercise support system comprising a hint information display unit displayed on a terminal for a user. An exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
The exercise support system includes an exercise management unit that sorts requests from learner terminals according to request types, a material database that stores multimedia data displayed on the learner terminals, and inputs from the user to the learner terminals. When a user assigns a material data storage unit that outputs material data in accordance with a requested request, a material data display unit that references material data on a learner's terminal, and an annotation associated with a location in the material data An annotation information input unit that generates annotation information in association with the annotation, an annotation information storage unit that receives the annotation information by the exercise support server, an annotation information database that stores the annotation information in association with a keyword, and aggregating annotation information , Your assigned keywords Annotation information totaling unit that aggregates the keywords, and when the user who requested the hint assigns the annotation, the keyword associated with the annotation appears less frequently than other users as a hint. An exercise support system comprising: a hint information generation unit that performs a hint information display unit that displays the generated hint information on a learner's terminal. An exercise support system in which a plurality of learner terminals and at least one teacher terminal are connected to an exercise support server via a network,
The exercise support system includes an exercise management unit that distributes requests from learner terminals and teacher terminals according to request types, a material database that stores multimedia data displayed on the learner terminals, and a user to a learner. A material data storage unit that outputs material data in accordance with a request input to the terminal for the user, a material data display unit that references the material data on the learner's terminal, and an annotation associated with the location in the material data by the user Annotation information input unit, an annotation information storage unit that receives the assigned annotation information by the exercise support server, an annotation information database that stores the assigned annotation information, and annotations that are stored. To classify according to location information The report summarizing section, the advice assisting information generating section that detects less than a certain number of annotated users from the annotation summing result and generating the corresponding section as advice assisting information, and displaying the advice assisting information on the teacher terminal An exercise support system characterized by having an advice auxiliary information display unit. An exercise support system in which a plurality of learner terminals and at least one teacher terminal are connected to an exercise support server via a network,
The exercise support system includes an exercise management unit that distributes requests from learner terminals and teacher terminals according to request types, a material database that stores multimedia data displayed on the learner terminals, and a user to a learner. A material data storage unit that outputs material data in accordance with a request input to the terminal for the user, a material data display unit that references the material data on the learner's terminal, and an annotation associated with the location in the material data by the user An annotation information input unit for generating annotation information in association with a keyword, an annotation information storage unit for receiving the annotation information by an exercise support server, an annotation information database for storing the annotation information in association with a keyword, and annotation information When counting An annotation information totaling unit that counts up for each keyword, an advice auxiliary information generating unit that detects fewer than a certain number of annotated users from the annotation totaling result, and generates the corresponding keyword as advice auxiliary information, and a teacher An exercise assistance system characterized by having an advice assistance information display unit for displaying advice assistance information on a terminal. An exercise support method using an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
When a material information acquisition request is received from a user, the corresponding material is searched from the material database, output and displayed on the user terminal,
Accepts annotations associated with locations in the material data,
Store the input annotation information in the annotation information database,
A practice support method for generating hint information using a result of the annotation information including the granting status of a plurality of users, and displaying the generated hint information on a learner's terminal that has requested the hint information output . An exercise support method using an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
When a material information acquisition request is received from a user, the corresponding material is searched from the material database, the material data is displayed on the screen of the learner's terminal,
Accepts annotations associated with locations in the material data,
Store the input annotation information in the annotation information database,
When a hint information output request is received from a user, the stored annotation information is detected by a location where a certain number or more of the users have annotated and the user who requested the hint does not give it. A practice support method that outputs information as information and displays the generated hint information on the learner's terminal. An exercise support method using an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
When a material information acquisition request is received from a user, the corresponding material is searched from the material database, the material data is displayed on the screen of the learner's terminal,
Accepts input of annotations associated with locations in the material data along with keywords,
Store the input annotation information in the annotation information database,
When a hint information output request is received from the user, the stored annotation information is aggregated for each keyword, and when the user who requested the hint assigns the annotation, the frequency of occurrence of the keyword associated with the annotation is different from the other. A practice support method for outputting keywords that are lower than a certain percentage as hint information in comparison with a user and displaying the generated hint information on the learner's terminal. An exercise support method using an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
When a material information acquisition request is received from a user, the corresponding material is searched from the material database, the material data is displayed on the screen of the learner's terminal,
Accepts annotations associated with locations in the material data,
Store the input annotation information in the annotation information database,
Upon receiving an advice auxiliary information output request, an exercise support method for detecting locations where the number of annotated users is less than a certain number, generating the corresponding locations as advice auxiliary information, and displaying the advice auxiliary information on the teacher terminal . An exercise support method using an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
When a material information acquisition request is received from a user, the corresponding material is searched from the material database, the material data is displayed on the screen of the learner's terminal,
Accepts input of annotations associated with locations in the material data along with keywords,
Store the input annotation information in the annotation information database,
Upon receiving an output request for advice assistance information, the annotation user detects fewer than a certain number of keywords from the annotation aggregation result, generates the corresponding keyword as advice assistance information, and the advice assistance information is sent to the teacher terminal. Exercise support method to be displayed. A program for exercise support in an exercise support system in which a plurality of user terminals are connected to an exercise support server via a network,
A computer constituting the exercise support server;
When receiving a material information acquisition request from the user terminal, a means for searching for the corresponding material from the material database and outputting and displaying it on the user terminal;
Means for receiving an annotation input associated with a location in the material data from a user terminal;
Means for storing the input annotation information in the annotation information database;
When a hint information output request from a user terminal is received, hint information is generated using the totaled result of the annotation information including the grant status of other users, and the generated hint information is learned. Means for displaying on the terminal for the user,
A program for supporting exercises to function as

Images