=Paper=
{{Paper
|id=Vol-1427/paper10
|storemode=property
|title=Morphological Localization Solution in e-Learning
|pdfUrl=https://ceur-ws.org/Vol-1427/paper10.pdf
|volume=Vol-1427
}}
==Morphological Localization Solution in e-Learning==
https://ceur-ws.org/Vol-1427/paper10.pdf
Morphological Localization Solution in e-Learning
Martin Jovanović Dejan Todosijević
Faculty of Electronic Engineering Niš Faculty of Electronic Engineering Niš
Aleksandra Medvedeva 14 Aleksandra Medvedeva 14
18000 Niš, Serbia 18000 Niš, Serbia
+381 18 529 523 +381 18 529 443
martin.jovanovic@elfak.ni.ac.rs todosijevic.dejan@gmail.com
ABSTRACT than technology, the spatial and often temporal distance between
This paper represents a follow-up to the first author's BCI 2013 the student and the teacher [2]. This approach opens up a vast
paper on the topic of localization in an e-learning framework and array of new possibilities, to the extent of an entire new learning
as such it discusses the implementation of the solution proposed paradigm – the one in which the driving force for learning shifts
therein. The motivation for these papers is the problem of from the teacher (the push approach) to the student (the pull
morphology in Western Balkans (or any other morphology- approach), as well as the quantity (the corporate-beloved "just
dependent) languages which imposes an additional layer of enough" principle) and the timing ("just in time" or "on demand"
complexity onto e-learning systems that provide any form of learning). [3] In order to provide required contents in a required
natural language feedback to users. One such system has been timeframe, large scale e-learning systems often store the learning
developed at Computer Science Department of the University of material in relatively small units ("learning objects") which are
Niš, Faculty of Electronic Engineering. This system provides combined and sequenced dynamically – often with the aid of other
elementary feedback in form of a relation between two arbitrarily technologies, like ontologies [4] or other Semantic Web tools. [5]
chosen notions in the learning material. In this contest the Semantic web turned out to be a perfect supplement to e-learning
returned statement, in the form of subject, predicate and object, [6] as it provides not only means for structuring, aggregation and
often requires a case of the object (noun) that is different from the organization of learning material, but also a better personalization
original case in the text. In languages that rely on morphology a of e-learning experience by providing means for student
different case means a different form of the word. Without the modeling. All this, as well as the high degree of scalability built in
proper case the returned statement is still understandable; into the core of Semantic web paradigm, drove the development
however, its improper grammar may be distractive to the learner. of the DSi e-learning framework, the central topic of this paper, to
This obstacle has been overcome in a relatively simple fashion by some of the Semantic web technologies. Though this framework is
additional semantics. This paper discusses the exact ways. in early stage of development and relatively simple in structure,
the choice of the Semantic web approach guarantees its
scalability.
Categories and Subject Descriptors
D.3.1 [Computer Uses in Education]: Computer-Assisted
Instruction (CAI) – Semantic Web, RDF and drag-and-drop user 2. THE DSI PROBLEM
interface in the context of learning The ACM Computing As the Drag-and-Drop Semantic Interface e-learning framework
Classification Scheme: http://www.acm.org/class/1998/ (DSi) has been thoroughly discussed at the BCI conference, [1] in
this paper only the brief review of its features will be given. The
framework has been prototyped [7] with the idea of high
General Terms scalability (by using Semantic Web tools) and graph-based
Algorithms, Design, Languages. knowledge representation (inspired by the idea of concept maps).
[8]
Keywords
e-learning, CAI, semantic web, instructional design, RDF, drag 2.1 DSi Framework Review
and drop, interface, DSi. The DSi framework consists of two layers: the textual layer
(learning material in plain or rich text) and the semantic layer (the
1. INTRODUCTION graph of notions from the text interconnected by relations between
Though the primary motivation for this paper and the underlying them). The latter is given in the form of RDF/XML. [9]
research is linguistic in nature, it is distraction-free e-learning that On page load, the text in not sent to the client immediately. It is
is in focus of both. The primary goal of e-learning is facilitation, first parsed for any words that exist in the semantic (RDF)
acceleration and enrichment of learning by the means of computer document. Once matches are found, all those words are
technology. Design of e-learning systems usually assumes, other encapsulated in span tags which are provided with drag and drop
capability. This way all the words that exist in the RDF document
can be dragged and dropped onto in the text displayed to the
Copyright © 2015 for the individual papers by the papers' authors. learner. This is the first important stage in the framework lifecycle
Copying permitted only for private and academic purposes. (Figure 1).
In: A. Bădică, M. Colhon (eds.): Proceedings of the 2015 Balkan When the user drags one word and drops it onto another (the word
Conference on Informatics: Advances in ICT can only be dropped onto another drag gable word, Figure 2), the
65
dragged and the dropped-onto word are sent to the server, where English learning text (as shown in Figures 1 and on). First time
the RDF document is searched for any RDF statement that Serbian language was applied, the morphology issue emerged: the
includes both of these words as statement subject and object, object word in a natural language sentence, almost always, has a
respectively. different lexical form in the subject-predicate-object (SPO)
statement than in the text. This is due to cases – in Serbian
language there are 7. In order to have a lexical match between the
word in the text and the word in the returned SPO statement, the
word in the text must (happen to) be in the same case as in the
SPO statement (in practice most frequently the 4th and 7th case). In
a free-style natural language text this may and may not happen
(the probability is, roughly, 1/7). The examples from [1] may
illustrate this more clearly. In Serbian, nouns come in 7 cases (the
word "stolica" – "chair" - is taken as an example):
Figure 1. Lecture text with draggable words. 1. nominative ("who?") - stolica,
2. genitive ("of ") - stolice,
3. dative ("Give to…") – stolici... etc.
Similar situation exists in Greek language, only there are 4 cases;
technically, the issue is identical (anthropos, Greek for man):
1. o anthropos,
2. tou anthropou... etc.
The word "stolica" (chair) can be found in any of 7 cases
throughout the text. However, in a relation such as "laptop is on
the chair" ("laptop je na stolici") requires the word "stolica" to be
Figure 2. Drag and drop operation. in the 7th case ("stolici"); there is no guarantee that this situation
will happen in the natural text and there is now way to force it.
If one or more such statements exist, the server will return Therefore, when a SPO statement is displayed to the user, it will
statement predicates in all of them. These predicates are relations most probably be grammatically incorrect.
between the two words user has chosen (Figure 3).
3. THE SOLUTION
This issue has been solved by deepening the semantic layer of the
document by one level. In order to describe this, a brief overview
of the semantic layer will be given.
3.1 State of Affairs on the Semantic Side
Each textual lesson in DSi framework is accompanied by an RDF
document. This document contains the mapping between certain
notions from the text (technically – relations between certain
words). For example, one RDF statement will define relations that
the word "Erickson" has with all the words it's related to. This is
Figure 3. Resulting statement (relation). best shown in the form of a graph (Figure 4).
In the RDF/XML syntax, this statement will look similarly to the
Figure 5, with possible slight variations in code.
The framework is, like mentioned before, in an early stage of
development, thus there is a lot of space for improvements. For Relations are stated in a simple form – as an element of an RDF
example, relations are given in a free, human language form, and statement (or part of it). In order to define the required case for
as such do not carry any formal semantics – prone to automated the object word in the statement, additional information must be
reasoning or classification. However, the primary intent of this provided in the RDF: each predicate's required object-word case.
version is testing on human subjects and gaining knowledge on In order to achieve this, compound RDF statements had to be
practical usability and initial shortcomings. One of these was the broken down into singular ones – so that each singular RDF
topic of [1], and another was encountered while the first one was statement (carrying only one relation) be enriched by this
overcome and tested. additional parameter – required object-word case.
2.2 Morphology Issue
The first usability issue was the question of morphological
transformations of nouns in morphology-dependent languages,
such as most of the Western Balkans languages. Though
development of the framework in being carried in Serbian
language environment, initial versions were implemented with the
66
2. Both words are turned into nominative (1st) case. This is
done by searching all words in all cases; once the match
is found, on any case, the nominative case of that word
is returned.
3. Based on both words in 1st case the predicate is sought
for (the same way as in the previous DSi version). [1]
4. Once the predicate is obtained, the adequate case for the
object-word is determined by the case requirement
statement (Figure 7).
5. Once the case is determined, the appropriate case of the
object word is obtained from the additional RDF
Figure 4. Relations in the RDF document. document which contains only cases for the words in
the primary RDF and all object-word candidates from
the text (one typical RDF statement from this document
Erickson Bandler
genitive
Figure 7. Case requirement statement.
Figure 5. RDF/XML syntax of a statement.
An example of the singular RDF statement, in the syntax of the Erickson
latest revision of the framework, could look like in Figure 6. To Ericksona
emphasize the morphological aspect, an inverse relation has been Ericksonu
given (instead of "Erickson was modeled by Bandler", the relation Ericksona
"Bandler modeled Erickson" was given as it provides a better Ericksone
example for morphology). Ericksonom
Ericksonu
Figure 8. A case RDF document statement.
Erickson
This way the statement returned to the user will always have a
Figure 6. A singular relation.
proper grammar. However, during the testing, another issue
In the initial framework version, when the word Bandler would be emerged.
dropped onto the word Erickson, the system would return the
statement "Bandler modeled Erickson". In Serbian, this would be: 4. ADDITIONAL PROBLEM
Bandler je_modelovao Erickson. In the initial version of the framework the learning text was in
English. In transition to Serbian, the case problem in the returned
This sentence requires the object word (Erickson) to be in the 4th statements was spotted and the solution was implemented. During
case. However, in the text, this might not be the case. So, if the further testing, a new issue emerged: in Serbian texts the
object-word is taken directly from the text, it will show in a wrong framework was unable to make all the words found in the RDF
case in the statement returned from the system. document draggable.
The initial solution proposed in [1] suggested that a new piece of
The sequence of the execution was the following:
information be added to each singular RDF statement: required
case. The case was to be given by its name, though number or any 1. split the text into separate words by traversing the DOM
ID would do. However, during the practical implementation tree and focusing on terminal nodes;
somewhat different solution was proved to be more adequate.
2. search the RDF document for each word from the text;
3.2 Solution Implementation 3. in case of a match, make the word draggable and
Instead of adding required case into the singular RDF statement, highlight it (with color).
the process which handles the drop operation took the different Not all the words that existed in the RDF document were
path: highlighted; even several instances of the same word were
1. When the word is dropped onto another, the two words highlighted and some were not. This occurred due to the lexical
(subject and predicate) are sent to the server. comparison with the words in the RDF – while all the words in
RDF statements are in the 1st case (nominative). Only nouns in the
67
1st case throughout the text were positively compared to the words forms an entirely new class of words to be addressed in further
in the RDF and thus marked as draggable. Words in any other research.
case (with the exception of some specific words) we false Other possible research directions have already been stated in [1]
negatives – the lexical comparison was negative due to the and include enriching the relations with properties (transitive,
morphological transformations in cases. reflexive or symmetrical), bringing relations themselves into
relations etc.
4.1 Solution
The solution required a change in the execution sequence and 6. REFERENCES
comparison. The new RDF document, containing cases for object- [1] Jovanović, M. - " Localization in E-learning Semantics (DSi
candidate words, needed to be expanded to encompass both Model Approach)", 6th Balkan Conference on Informatics
subject and object candidates. BCI 13, Thessaloniki, Greece, September 19-21, 2013,
On page load, any word from the text is compared against all Proceedings of the 6th Balkan Conference in Informatics, pp.
cases of both subject and object candidate words; that part the 164-170, published by ACM New York, NY, USA ©2013,
execution sequence is altered (the system addresses the case RDF ISBN: 978-1-4503-1851-8
document instead of the relations RDF document). In this [2] Ally, M. – Foundations of Educational Theory for Online
implementation all instances of all subject-candidate and object- Learning, u Anderson, T. (editor) - The Theory and Practice
candidate words were recognized, highlighted and spanned as of Online Learning, Athabasca University Press, 2008, ISBN
drag-droppable. 1897425082, pp. 15-44.
This aspect was completely neglected when the case problem was [3] Horton, W. – Leading E-Learning: Here is how You Can
first, tackled and it wasn't until the first practical testing of the Chart your Course, Champion Implementation, Ensure
solution proposed in [1] that the attention was drawn to it. Success, American Society for Training and Development,
2001, ISBN 1562862987.
5. CONCLUSION [4] Fayed G., Sameh D., Ahmad H., Jihad A., Samir E. i Hosam
This paper discussed the solution to the localization problem in an E. - E-Learning Model Based On Semantic Web Technology,
e-learning system aimed at morphology-rich languages, such as International Journal of Computing & Information Sciences,
Western Balkans region languages. The e-learning framework in Vol. 4, No. 2, 2006, pp. 63–71.
question is labeled DSi (Drag and Drop Semantic Interface) [10] [5] Semantic Web, W3C Standards Web page, accessible online
and provides means to retrieve relations between any two words at: http://www.w3.org/standards/semanticweb/ (URL last
in the text on the go, as the learning progresses, with no need for st
visited on March 31 , 2015).
reviewing definitions at the beginning of the course. The
framework is operational for learning materials in English or any [6] Pahl, C., Holohan, E.: Applications of Semantic Web
other language with morphologically stable nouns. However, in Technology to Support Learning Content Development,
languages that rely on morphology to convey the meaning (in Interdisciplinary Journal of E-Learning and Learning
which nouns morphologically alter depending on the context), the Objects, vol. 5, 2009.
framework returns grammatically improper responses. The [7] Jovanović, M. - "Arhitektura konceptualno-orijentisanog
solution was proposed in [1] and implemented with some alata za podršku učenju", Infoteh Conference, Jahorina,
differences to the proposed way. Testing the solution revealed Republic od Srpska, 28-30. 03. 2007, Infoteh-Jahorina
another issue of the exact same nature which caused the problems Proceedings of Papers, Vol. 6, Ref. E-IV-4, pp. 477-479,
on an unexpected location – in the initial application loading published by Faculty of Electronic Eastern Sarajevo, R.
phase, limiting the framework functionality (a more serious Srpska, 2007, ISBN 99938-624-2-8.
limitation than the improper grammar of the responses). This issue
led to some framework implementation changes, generalizing the [8] Novak, J. D. & A. J. Cañas, The Theory Underlying Concept
case RDF document application to both application load phase Maps and How to Construct Them, Technical Report IHMC
and user interaction (drag and drop action) handling. CmapTools 2006-01 Rev 01-2008, Florida Institute for
Human and Machine Cognition, 2008.
Completing the application in this respect hasn't, however,
resolved all the potential morphology-induced issues. Though [9] RDF (Resource Description Framework) - W3C
fully operational for the current purposes, [11] the application Specification available at: http://www.w3.org/RDF/ (URL
might not operate properly with other morphological last accessed on 1st of April, 2015.
transformations, such as possession. This aspect is relatively easy [10] Jovanović, M. - "Semantička nadgradnja nastavnog
to address in English (where possessions include "'s" suffix which materijala u sistemima za elektronsko učenje" ("Semantic
can be tackled with a regular expression). However, possessions Upgrade of Learning Material in E-Learning Systems"),
in Serbian are built morphologically (and as such can serve as M.Sc. thesis, University of Niš, Serbia, Faculty of Electronic
nouns) and are also prone to case changes. Example would be Engineering, 2009.
"Martin's" which translates to "Martinov" for male, "Martinova" [11] The DSi showcase application can be found at the following
for female and "Martinovo" for neutral possession, all of which URL: http://www.martin.rs/DSi1.5_TestModule_v4/. The
come in 7 forms (cases) and can, in certain situations, assume the URL is prone to change. In case of 404, please contact the
role of nouns (thus be candidates for dragging and dropping). This authors.
68