TWI797541B - Interactive record parsing method - Google Patents

Abstract

An interactive record parsing method includes steps of providing a record database and an interactive record parser including a user interface, allowing the interactive record parser to parse a plurality of records of the record database and display a parsing result on the user interface, building at least a rule, allowing the interactive record parser to apply the rule to parse the records, and extracting at least an event. As a result, not only the events of the records can be precisely reflected, but also behaviors and related knowledges existing in the records can be provided, thereby achieving the advantages of facilitating research.

Description

Interactive record parsing method

This case relates to a method of parsing, especially an interactive record parsing method.

With the gradual popularization of computer-based learning environments, computer systems automatically generate records to track user activities and behaviors, allowing researchers to track learning events without interrupting learners' thinking and browsing. More importantly, this method can virtually reproduce the behavior of learners to further understand how learners construct knowledge and use the system. In e-learning, the extraction of sequences and behavioral patterns is helpful for educational research, such as evaluating learners' activities and comparing them with expected behavioral patterns to identify factors or patterns that contribute to success.

Behavior analysis is the basis of sequence and behavior pattern analysis, and its purpose is to extract learners' behavior according to characteristics or detailed behavior sequences. In general, logs of learner interactions with the system or dedicated recording applications or language can be used for behavioral analysis. Extracting and analyzing learners' behavior through statistical techniques is the most common practice, but it often only reflects the learners' behavior itself, and cannot provide relevant knowledge existing in the log. In addition, a single action performed by a learner may be recorded by multiple logs. However, the existing statistical techniques cannot effectively detect or integrate the learner's operations in multiple logs before analysis, and cannot fully express the richness of learner interaction. .

To put it simply, given the richness of e-learning and user interaction, existing technologies cannot meet the needs of effectively analyzing interactive behavior. In the hierarchical structure of e-learning, similar interactions may have different learner behaviors due to the different learning activities involved, but existing technologies cannot fully analyze and reflect this complex hierarchical relationship. For example, different orders of logging may achieve the same behavior; or in different tasks or events, the same order of logging may represent different behaviors. Clearly, existing record parsing techniques are not flexible enough. On the other hand, due to possible errors in the system or network delays, there may be problems such as repeated records or partial records being omitted and out of sequence in the logs. However, the existing technology does not have a dedicated interface to view and organize logs and adjust rules. Record parsing is messy and unwieldy.

Therefore, how to develop an interactive record analysis method that can effectively solve the problems and shortcomings of the aforementioned prior art is an unresolved problem.

The main purpose of this case is to provide an interactive record analysis method to solve and improve the problems and shortcomings of the aforementioned prior art.

Another purpose of this case is to provide an interactive record analysis method. The interactive record parser parses the records and displays the analysis results on the user interface, and then creates rules and applies them to the records to extract events, which can not only accurately reflect the events in the records , can also provide the behavior and related knowledge existing in the records, and then achieve the effect of facilitating research.

Another purpose of this case is to provide an interactive record analysis method. By continuously judging extraction requirements and establishing tokenization rules, and applying test content to continuously analyze and monitor results, it can effectively meet the needs of interactive behavior in analyzing events, and can target different The event level can effectively analyze and classify the actual behavior, and then it can deal with complex records and show the richness of e-learning and user interaction. At the same time, it can detect errors in the records and display them in the user interface by category , to effectively simplify recording and analysis.

In order to achieve the above purpose, one of the preferred implementation forms of this case is to provide an interactive record analysis method, including the steps of: (a) providing a record storage library and an interactive record parser, wherein the interactive record parser includes a a user interface; (b) allow the interactive record parser to parse a plurality of records in the record repository and display a result of the analysis in the user interface; (c) create at least one rule; (d) the interactive A formula record parser applies the rule to parse the plurality of records; and (e) extracts at least one event.

In order to achieve the above purpose, one of the preferred implementation forms of this case is to provide an interactive record analysis method, including the steps of: (a) providing a rule database, a record resolver and a user interface; (b) allowing the use of interacting with the record parser and the rule database to create at least one tokenization rule, store the tokenization rule in the rule database, apply the tokenization rule to the record parser, and through the use (c) input a plurality of records to the record parser; and (d) the record parser parses the plurality of records according to the tokenization rule to extract at least one event.

Some typical embodiments embodying the features and advantages of the present application will be described in detail in the description in the following paragraphs. It should be understood that this case can have various changes in different aspects, all of which do not depart from the scope of this case, and the descriptions and diagrams therein are used as illustrations in nature, rather than construed to limit this case.

Please refer to FIG. 1, which is a flow chart showing an interactive record analysis method according to an embodiment of the present case. As shown in FIG. 1, the interactive record analysis method of an embodiment of the present case includes the following steps. Firstly, as shown in step S10, a record storage library and an interactive record parser are provided, wherein the interactive record parser includes a user interface (User Interface). Secondly, as shown in step S20, the interactive record parser is allowed to parse multiple records in the record storage, and display the analysis results on the user interface. It should be noted that the multiple records described here can be obtained from The sample records (Sample Logs) selected from the record storage repository. Next, as shown in step S30, at least one rule is established. In some embodiments, this step S30 may include a sub-step of storing the rule in a rule database, but not limited thereto. Then, as shown in step S40, the interactive record parser applies rules to parse the plurality of records. Next, as shown in step S50, at least one event is extracted. It should be noted that the records (Logs) mentioned in this paragraph may refer to records of instructions in a system, and in a learning system may be records of interactions between learners and the learning system, but are not limited thereto; The event mentioned in the paragraph can be a simple behavior or action, such as watching a movie, clicking a button or ending the study, etc., or it can be the learning content of a unit or a complete chapter, but it is not limited to this . The interactive record parser parses the records and displays the analysis results on the user interface, and then creates rules and applies them to the records to extract events. It can not only reflect the events in the records, but also provide the behaviors and related knowledge existing in the records. In order to achieve the effect of facilitating research.

Please refer to FIG. 2 , which is a flowchart showing the sub-steps of step S20 of the interactive record analysis method of an embodiment of the present case. As shown in Figure 2, step S20 of the interactive record analysis method in this case includes sub-step S24 and sub-step S26, wherein sub-step S24 identifies multiple attributes of multiple records, and sub-step S26 identifies the multiple attributes corresponding to multiple attribute names and plural attribute values into plural simplified explanations. That is to say, in step S20, when the interactive record parser parses the multiple records in the record storage repository, the interactive record parser recognizes the multiple attributes of the multiple records, such as user name, operation instruction, location Sections, lessons, chapters, activities and actions, and more. Then, the interactive record parser converts the corresponding plurality of attribute names and plurality of attribute values into a plurality of simplified explanations.

Please refer to FIG. 3 , which is a flowchart showing the sub-steps of step S20 of the interactive record analysis method of an embodiment of the present case. As shown in Figure 3, step S20 of the interactive record analysis method in this case, after sub-step S24 and sub-step S26, further includes sub-step S28, which integrates multiple records and multiple simplified explanations corresponding to multiple records into Steps to parse the results. To put it simply, the interactive record parser can simplify and integrate multiple records into analysis results, which can be displayed on the user interface for users to watch. It should be noted that the users mentioned here may be the researchers recorded in the research record keeping database, but generally not the learners mentioned in the preceding paragraphs.

Please refer to FIG. 4 , which is a flowchart showing the sub-steps of step S20 of the interactive record analysis method according to an embodiment of the present case. As shown in FIG. 4, the step S20 of the interactive record analysis method of this case may further include a sub-step S22 before the sub-step S24, that is, a step for the interactive record parser to obtain the event list. The event list preferably enumerates all possible events, which can assist the interactive record parser to parse multiple records more accurately, but is not limited thereto.

Please refer to Figure 5 and cooperate with Figure 1, where Figure 5 is a schematic diagram showing the establishment of a tokenization rule by the interactive record analysis method of this case. As shown in Figures 1 and 5, in the interactive record analysis method of this case, each rule is preferably a tokenization rule (Tokenization Rule), and the rule database is preferably a tokenization rule database, However, it is not limited to this. In addition, in step S30, preferably, the interactive record parser establishes at least one tokenization rule according to a plurality of simplified interpretations. To put it simply, when the interactive record analysis method of this case is applied to a system, such as a learning system, tokenization rules can be established according to the record format and logic stored in the system, such as extracting the region in a record segment start message S11 and segment end message S12, and establish a marking rule. Then, apply this tokenization rule to multiple records, all corresponding segment start messages S11 and segment end messages S12 can be extracted from multiple records, and integrated into a single segment marker S1 for display in use interface.

Similarly, the interactive record parsing method in this case can also extract the chapter start message C11 and chapter end message C12 from a record, and create a tokenization rule. Then, by applying the tokenization rule to multiple records, all the corresponding chapter start messages C11 and chapter end messages C12 can be extracted from the multiple records, and integrated into a single chapter marker C1 for display on the user interface.

Of course, the interactive record analysis method in this case can also use a similar method to extract the section start message S21, section end message S22, chapter start message C21, chapter end message C22, title start message T11, and title end message T12, and are respectively integrated into the section mark S2, the chapter mark C2 and the title mark T1, which will not be repeated here. It is worth mentioning that whether it is a section mark S1, or a section mark S2, a chapter mark C1, a chapter mark C2, and a title mark T1, it can be regarded as an event, that is to say, in step S50, according to the mark The optimization rule extracts at least one event from the plurality of records.

Of course, the events mentioned are not limited to the above-mentioned examples, and may also include all possible events in different systems. To put it simply, the interactive record analysis method in this case can be easily and systematically applied to various types of systems, and its generalization ability is good, which can cope with different logics and record formats of various systems, and provides users with The interface is for users to operate and browse, and has good interactivity.

Please refer to FIG. 6, which is a flowchart showing an interactive record analysis method according to an embodiment of the present case. As shown in FIG. 6 , the interactive record analysis method according to an embodiment of the present case may include step S60 after step S50 , which is a step of judging whether the event satisfies the extraction requirement. In some embodiments, each rule has a corresponding at least one extraction requirement. When the judgment result of step S60 is negative, step S20 is re-executed after step S60, that is, re-analyzing and establishing rules and corresponding and appropriate extraction requirements. When the judgment result of step S60 is yes, the following steps are executed after step S60: as shown in step S70, a plurality of important elements are selected as test content. Next, as shown in step S80, the record keeping library is continuously analyzed and the results are monitored. Then, as shown in step S90, it is judged whether an error occurs. When the judgment result of step S90 is yes, step S20 is re-executed after step S90, and when the judgment result of step S90 is no, step S80 is re-executed after step S90. Among them, important elements can be selected according to actual needs as test content to test whether the rules established in the preceding steps can be effectively used to analyze all records in the record storage library.

From the above description, it can be understood that the interactive record analysis method in this case can effectively meet the needs of analyzing interactive behavior in events by continuously judging extraction requirements and establishing tokenization rules, and applying test content to continuously analyze and monitor results. The event level can effectively analyze and classify the actual behavior, and then it can deal with complex records and show the richness of e-learning and user interaction. At the same time, it can detect errors in the records and display them in the user interface by category , to effectively simplify recording and analysis.

In some embodiments, step S30 is implemented by the user through the user interface, step S50 is implemented by the interactive record parser, and step S60 is automatically implemented by the interactive record parser or by the user through the user interface realized, but not limited to. In some embodiments, the user interface is provided for users to browse rules, create rules, browse extraction requirements, create extraction requirements, and modify extraction requirements, etc., but it is not limited thereto.

Please refer to FIG. 7, which is a flowchart showing an interactive record analysis method according to an embodiment of the present case. As shown in Figure 7, steps S10 to S60 of the interactive record analysis method in this case can be regarded as a record analysis and testing platform 100, whose functions include selecting and browsing records and subsets of records, creating and analyzing rules, and viewing processed data, etc. On the other hand, the steps S70 to S90 of the interactive record analysis method in this case can be regarded as the record analysis monitoring and validator 200, which includes a record analysis monitoring function and a record analysis verification function, and the record analysis monitoring function can apply the created rules to A complete record keeping repository, monitoring event extraction results and errors, and feeding back to the record parsing and testing platform 100 to fine-tune rules. The record parsing and verification function can continuously confirm whether the processed records are normal, and whether the verification rules or extracted events can effectively correspond to new data or new features in the input records.

Based on the aforementioned framework, this case also provides an interactive record analysis method. Please refer to Figures 8 and 9, wherein Figure 8 is a flow chart showing an interactive record analysis method according to an embodiment of this case, and Figure 9 is a diagram showing the architecture of an interactive record analysis system according to an embodiment of this case. As shown in FIG. 8 and FIG. 9, the interactive record analysis method of an embodiment of the present case includes the following steps. First, as shown in step S100 , a rule database 11 , a record parser 12 and a user interface 13 are provided. Next, as shown in step S200, the user interface 13 is allowed to interact with the record parser 12 and the rule database 11 to create at least one tokenization rule, store the tokenization rule in the rule database 11, and apply the tokenization rule to Record the parser 12, and monitor the operation of the record parser 12 through the user interface 13. Then, as shown in step S300 , a plurality of records are input to the record parser 12 . Next, as shown in step S400 , the record parser 12 parses the plurality of records according to the tokenization rules to extract at least one event. In this way, not only can the events in the records be accurately reflected, but also the behaviors and related knowledge existing in the records can be provided, thereby achieving the effect of facilitating research.

Based on the aforementioned interactive record analysis method in this case, interactive models corresponding to various types of systems can be easily established. Please refer to Figure 10, which is a schematic diagram showing the establishment of an interactive model through the interactive record parser of this case. As shown in Figure 10, when there are interaction records or interaction logs in any system, the interactive record analysis method and interactive record parser of this case can be applied to analyze the interaction records or interaction logs and extract events, and then gradually build Generate an interaction model applicable to the current system.

To sum up, this case provides an interactive record analysis method. The interactive record parser parses the records and displays the analysis results on the user interface, and then creates rules and applies them to the records to extract events. Not only can it truly reflect the events in the records , can also provide the behavior and related knowledge existing in the records, and then achieve the effect of facilitating research. In addition, by continuously judging and extracting requirements and establishing tokenization rules, as well as continuously analyzing and monitoring results by applying test content, it can effectively meet the needs of analyzing interactive behaviors in events, and can effectively analyze actual behaviors and analyze them for different event levels. Classification can deal with complex records and show the richness of e-learning and user interaction. At the same time, errors in records can be detected and displayed on the user interface by category, effectively simplifying records and analysis.

Even though the present invention has been described in detail by the above-mentioned embodiments and can be modified in various ways by those who are familiar with the art, they are all within the scope of protection intended by the appended patent application.

11: Rule Database 12:Record Parser 13: User Interface 100: Record analysis and test platform 200: Record parsing monitor and validator C1: Chapter markers C11: Chapter start message C12: Chapter end message C2: Chapter markers C21: Chapter start message C22: Chapter end message S1: Segment mark S11: Section start message S12: Section end message S2: Segment mark S21: Section start message S22: Section end message S10: step S20: step S22: sub-step S24: sub-step S26: sub-step S28: sub-step S30: step S40: step S50: Steps S60: Steps S70: Steps S80: step S90: Steps T1: title tag T11: Title start message T12: Title end message

Fig. 1 is a flow chart showing an interactive record analysis method according to an embodiment of the present case. FIG. 2 is a flow chart showing the sub-steps of step S20 of the interactive record analysis method according to an embodiment of the present case. FIG. 3 is a flowchart showing the sub-steps of step S20 of the interactive record analysis method according to an embodiment of the present case. FIG. 4 is a flow chart showing the sub-steps of step S20 of the interactive record analysis method according to an embodiment of the present case. Figure 5 is a schematic diagram showing the establishment of a tokenization rule by the interactive record analysis method of this case. Fig. 6 is a flowchart showing an interactive record analysis method according to an embodiment of the present case. Fig. 7 is a flowchart showing an interactive record analysis method according to an embodiment of the present case. Fig. 8 is a flowchart showing an interactive record analysis method according to an embodiment of the present case. Fig. 9 is a structural diagram showing an interactive record analysis system according to an embodiment of the present case. Figure 10 is a schematic diagram showing the establishment of an interactive model through the interactive record parser of this case.

S10: step

S20: step

S30: step

S40: step

S50: Steps

Claims (12)

A method of interactive record parsing comprising the steps of: (a) providing a record repository and an interactive record parser, wherein the interactive record parser includes a user interface; (b) allowing the interactive record parser to parse The plurality of records in the record storage repository, and display a resolution result on the user interface; (c) create at least one rule through the user interface; (d) the interactive record parser applies the rule to resolve the plurality records; and (e) the interactive record parser extracts at least one event from the plurality of records; wherein the step (b) includes the sub-steps of: (b1) identifying a plurality of attributes of the plurality of records; (b2) converting the plurality of attribute names and the plurality of attribute values corresponding to the plurality of attributes into a plurality of simplified explanations; and (b3) integrating the plurality of records and the plurality of simplified explanations corresponding to the plurality of records into the analysis result. The interactive record parsing method as described in Claim 1, wherein each of the rules is a tokenization rule, and the step (c) is to establish at least one of the tokenization rules according to the plurality of simplified explanations. The interactive record parsing method as described in Claim 2, wherein the step (e) is to extract at least the event from the plurality of records according to the tokenization rule. The interactive record parsing method as described in Claim 1, wherein the step (b) is before the sub-step (b1), and further includes a sub-step: (b0) the interactive record parser obtains an event list. The interactive record parsing method as described in claim 1, further comprising a step after the step (e): (f) judging whether the event satisfies an extraction requirement. In the interactive record parsing method as described in claim 5, each of the rules has at least the corresponding extraction requirement. The interactive record analysis method as described in Claim 5, wherein when the judgment result of the step (f) is negative, the step (b) is re-executed after the step (f). The interactive record analysis method as described in Claim 5, wherein when the judgment result of the step (f) is yes, the step (g) is performed after the step (f): (g) selecting a plurality of important elements as a test content; (h) continuously analyze the record storage and monitor the results; and (i) judge whether there is an error; wherein, when the judgment result of the step (i) is yes, the step (i) is re-executed after the step ( b), and when the judgment result of the step (i) is no, the step (h) is re-executed after the step (i). In the interactive record parsing method described in Claim 5, the step (f) is realized automatically by the interactive record parser or by the user through the user interface. The interactive record parsing method as described in Claim 5, wherein the user interface is provided for the user to browse the rule, create the rule, browse the extraction request, create the extraction request, and modify the extraction request. The interactive record analysis method as described in Claim 1, wherein the step (c) further includes a sub-step: (c1) storing the rule in a rule database. In the interactive record parsing method as described in claim 11, each of the rules is a tokenization rule, and the rule database is a tokenization rule database.

Images