1 / 17

Frame-Based Semantic Retrieval of Education Supportive Materials

Frame-Based Semantic Retrieval of Education Supportive Materials بازیابی معنایی مواد پشتیبان در آموزش بر پایه روش ساخت یافته (قاب). Maryam Tayefeh Mahmoudi School of ECE, College of Engineering University of Tehran, Iran tayefeh@ut.ac.ir. 2. Introduction. Old Educational Systems

turner
Download Presentation

Frame-Based Semantic Retrieval of Education Supportive Materials

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Frame-Based Semantic Retrieval of Education Supportive Materials بازیابی معنایی مواد پشتیبان در آموزش بر پایه روش ساخت یافته (قاب) Maryam Tayefeh Mahmoudi School of ECE, College of Engineering University of Tehran, Iran tayefeh@ut.ac.ir

  2. 2 Introduction • Old Educational Systems • New Educational Systems • Increasing digital resources including educational materials has made a great challenge for innovating techniques to solve the problem of finding the truly-necessary information.

  3. 3 Prof. Dr. Fred Mulder UNESCO Chair in OER at the Open University in The Netherlands Opening up Education آزاد/باز کردن آموزش “As institutions and as individuals, we seem to have forgotten the core values of education: sharing, giving, and generosity.”

  4. 4 Two Distinct Open Education Worlds 6-fold Classical Openness Open Access Freedom of Time Freedom of Pace Freedom of Place Open Programming Open to People 4-fold Digital Openness Family regarding free online availability: Open Source (software) Open Access (scientific output) Open Content (creative output) Open Educational Resources / OER (learning materials)

  5. 5 From OER to Open Education (OE) “... are teaching, learning and research materials in any medium that reside in the public domain and have been released under an open licence that permits access, use, repurposing, reuse and redistribution by others with no or limited restrictions.” (UNESCO 2011) OER ≠ (Open) Education Open Teaching Effort (OTE) Complementary to OER and OLS, to be paid for, referring to the human effort in different roles: developing, presenting, explaining, assessing, communicating, interacting, intervening, mediating, etcetera of teachers and educators (and with the learners in their specific role) in a professional, open, and flexible learning environment and culture. Open Learning Services (OLS) Complementary to OER, free or to be paid, and including a variety of online and virtual facilities for: tutoring, advice, meetings, communities, teamwork, presentations, testing, examination, consulting sources, internet navigation, etcetera …

  6. 6

  7. 7 Introduction • Many annotating, tagging and indexing techniques (statistical & semantical techniques) have been created to ease the way that learners get their appropriate resources. • They generally make use of ontology, WordNet and meta-data to describe the resources. • Latent semantic indexing (LSI) and concept indexing (CI) are the techniques capable of organizing educational assets on the basis of mapping between tagging and querying vocabularies. • Despite the vantages & effectiveness of these techniques within certain domains, they have their own limits and deficiencies in organizing resources with regard to inadequacy of reusability for support purposes. • A frame-based semantic technique is proposed to enhance the retrieval mechanism for educational resources usingtwo attributes called "Major Characteristics" and "Basic Constituents" which respectively stand for "the goal behind a concept" and "the elements that support a concept to be realized“. • It is not only capable of clarifying the existing ambiguity in tags, but also can embed the focal knowledge for exploring the similarity relevance between query and supportive materials.

  8. Insert a Query Extracting query's MJ & BC automatically Identifying Basic Constitute Part; { Applying Part of Speech Tagging; Analyzing Prepositions; Identifying Basic Constituents; } Rules Identifying Major Characteristics; { Determining the layers of MJ by analyzing prepositions; Determining the grammatical role of each layer by POS tagger; { Identifying Action Part; Identifying Adverb Part; Identifying Direct Object Part; Identifying Indirect Object Part; } } POS Tagger Data Base Search in DB of supporting materials based on MJ & BC N Search for synonyms of Query's terms in WordNet Found it? Y WordNet Return the corresponding material 8 Framework for Semantic Retrieval • Frames: determining grammatical roles of the existing terms in queries and titles with regard to education supportive materials in a database. • Two significant attributes are "Major Characteristics" and" Basic Constituents". • "MJ" stands for the main objective behind using a material. • "BC“ focuses on the methods, techniques or tools which are used to reach this objective

  9. 9 Frame’s Attributes is a noun or a pronoun that becomes subject to this verb or shows the result of the related action. It is able to answer "What?"s or "Whom?"s relating to this verb MJ" stands for the main objective behind using a material. Certain conjunctions and prepositions can be in charge of specifying MJs and BCs like "in" and "for" followed by a verb usually yield two layers for MJ, which follow the same structure including four main parts of "Action", "Adverb/Adjective", "Direct object" and "Indirect object”. "Considering agent mobility architecture for controlling transportation based on FIPA standards" Major Characteristics (MJ) is also the recipient of the direct object and has the ability to answer "To whom?"s or "For whom?"s and it usually follows a preposition.. which can modify verbs, adjectives, clauses, sentences, and other adverbs. It typically answers "How?"s, "In what way?"s, "When?"s, "Where?"s, and "To what extent"s, etc. is mainly a verb

  10. 10 Frame’s Attributes "BC“ focuses on the methods, techniques or tools which are used to reach this objective. For BCs, most of the time, one layer at maximum seems to be sufficient. Determining BC is closely related to the conjunctions which are considered for this purpose. Some of these conjunctions are "based on", "on the basis of", "on the ground of", "using", "making use of", “via” etc . "Considering agent mobility architecture for controlling transportation based on FIPA standards" Basic Constituents (BC) Having reviewed large amount of titles, several semantic rules are yielded for distinguishing MJs and BCs, For example: - The "word" or "phrase" coming after "via" or "based on" is most probably a BC. - The rest of the title consisting of a verb coming after "for" is most probably a MJ's 2nd layer.

  11. 11 How the Framework Can be Implemented?! Inserting a query; Parsing the query; Extract (MJ, BC) of query; { Identifying Basic Constitute Part; { Applying Part of Speech Tagging; Analyzing Prepositions; Identifying Basic Constituents; } Identifying Major Characteristics; { Determining the layers of MJ by analyzing prepositions; Determining the grammatical role of each layer by POS tagger; { Identifying Action Part; Identifying Adverb Part; Identifying Direct Object Part; Identifying Indirect Object Part; } } } Search in Database of supporting materials' tiltes (Mj, BC); If (query (Mj, BC) = supporting materials' tiltes (Mj, BC)); Send existing response based on relatedness of query (MJ, BC) to title (MJ, BC); Else Finding synonyms of terms of query in WordNet; Repeat the whole process to extract appropriate supporting materials for that; To apply frames to headers, first we split the header into "Basic Constituents" and "Major Characteristics". Subsequently, we determine whether the "Major Characteristics" hold one layer or two layers. Here, it is essential to find out the grammatical role of the terms included in the "Major Characteristics" as well as the "Basic Constituents".

  12. 12 Grammatical Rules to Determine the Attributes To realize the grammatical roles of the terms we may use of rules to decide what role a term can hold. For instance, to realize an "action-part" (in "Major Characteristics") rules can take into account information regarding suffixes like "tion", "sion", "ment", "ing", etc. which are linguistically significant. Let say, for example, if POS tagger is tagging a "tion" word without “tion” as a verb, and that word is not located between two nouns; one can conclude that the term must be an action. "Simulation of Dialogue Management" Simulation Simulate  is a verb Not between two nouns  is an action "Reliable transactions in multi-agent systems" Transaction Transact  is a verb Not between two nouns  is an action-type

  13. 13 Results of Some Experimentations We made a data set including 134 supporting materials in the domain of Agent Science and Technology. To perform our tests, we designed some questions and expected our approach to findproper materials as the answers of these questions. Results were evaluated through comparing the responses made by our approach with the real responses obtained from the experts of Agent Science & Technology. In fact out of 134 titles used in our experiments, the system has been able to function properly in 107 cases. This means that the precision in detecting "MJ"s and "BC"s is 93%, while the recall is 79%. “Precision” is calculated as the proportion of relevant retrieved documents to the number of retrieved documents. Precision = (31/33) = 0.93 “Recall” is calculated as the proportion of relevant retrieved documents to the total number of relevant documents. Recall=31/ (31+8) =0.79 F-measure = 2 * (Precision * Recall) / (Precision + Recall) =2 * (0.93 * 0.79) / (0.93 + 0.79) = 0.85 Precision & Recall of the proposed approach

  14. Precision & Recall of Some Queries 14

  15. 15 Future Prospects • A Frame-based semantic framework for retrieving education supportive materials is proposed. • Attributes called "Major Characteristics" and "Basic Constituents" can be used to realize semantic retrieval of education supportive materials in an efficient way. • These attributes were shown to be enough potential for representing the knowledge of titles and sub-titles in such a way reflecting paragraphs content in a reasonable way. • Some rules based on proposition and conjunctions, which are grammatically significant, were used to detect "BC"s and "MJ"s in the titles and subtitles. • Rules can be constituted based on the existing linguistic knowledge. The high number of proposition in a rule, a higher expectation would exist its effective role in detection. • For the moment to avoid extra computation, rules have been decided to include only a few predicates. However, developing more potential rules through considering further predicates can be regarded as a major research work for future.

  16. Thank you very much for your notice Thank you very much for your notice E-learning Laboratory University of Tehran www.telab.ut.ac.ir

More Related