Enhancing Web Document Accessibility by Authoring ... - Springer Link

Enhancing Web Document Accessibility by Authoring Texts and Text Comprehension Activities Grammatiki Tsaganou1, Maria Samarakou2, Panagiotis Blitsas3, and Maria Grigoriadou1 1

University of Athens, Dept. of Informatics & Telecommunications, Ilisia Panepisthmiopolis GR-15784, Athens, Greece {Gram, gregor}@di.uoa.gr 2 Technological Educational Institute of Athens, Department of Energy, Ag. Spyridonos Str. GR-12210, Egaleo, Athens, Greece [email protected] 3 University of Athens, I.P.G.S of Basic & Applied Cognitive Science, Ilisia Panepisthmiopolis GR-15784, Athens, Greece [email protected]

Abstract. In this paper we discuss recent directions concerning the structural analysis of science documents and cognitive aspects of document elements aiming at document comprehension. Structural analysis of documents, according to text comprehension theory, promotes document understanding and enhances universal accessibility of documents. We outline the process of structuring science documents and activities for comprehension using the authoring tool ReTuDiSAuth. This process improves universal accessibility of document supporting authors to structure text and text activities for students with different abilities, requirements and preferences. Keywords: Web document accessibility, structural analysis of documents, universal accessibility of documents, web authoring tools.

1 Introduction Universal accessibility of documents implies the global requirement for access by individuals with different abilities, requirements and preferences [18]. The process of integrating accessibility into texts and text activities accessed by the web, aims at facilitating a better understanding of texts. The cognitive psychological approach in text understanding suggests that the internal variables of the reader hold a primary role in text comprehension, such as his personal goals, interests and pre-existing knowledge. However, cognitive science does not ignore the influence of the text form, in which factors such as text cohesion and logical coherence of facts presented have been proved to be significant elements that facilitate its comprehension [7]. Many studies on text comprehension have focused their interest on the sentence structure presented by the text [5], [12]. Sentence structure of a text could be organized on the basis of hierarchy in order to allow the importance of sentences in the text to be C. Stephanidis (Ed.): Universal Access in HCI, Part III, HCII 2009, LNCS 5616, pp. 597–606, 2009. © Springer-Verlag Berlin Heidelberg 2009

598

G. Tsaganou et al.

revealed [22]. In approaching text comprehension, researchers examine issues that focus on assisting comprehension through text summarization [5] by improving text coherence [8], [12], [14] or assisting the design of the text form and text activities [3]. In order to make the information in such activities available to target users (students, teachers, researchers, authors, educators) new efforts have emerged to bring together novel methodologies and technologies. Authoring such activities demands an authoring system which involves knowledge acquisition, design process and managing a large amount of complex information. Authoring tools offer the appropriate structure and guide authors to import and elaborate educational material (text, activities, questions, dialogues etc.). Researchers have been investigated Intelligent Tutoring Systems (ITS) with authoring tools almost since the beginning of ITS research and authoring systems have been built [13], [17]. An authoring tool is a generalized framework along with a user interface that allows non programmers to formalize their knowledge [13], [16]. Part of authoring an ITS is the systematic decomposition of the subject matter into a set of related text elements. Each authoring system provides tools or cues which assist the author in this process of breaking down and elaborating the content to the necessary level of detail according to an instructional model. There are intelligent adaptive hypermedia systems like CALAT [15] and GETMAS (Wong & Chan, 1997) that their functions overlap those from both the above categories. There are also expert systems like IRIS-tutor [2], which include rule-based cognitive models of problem solving expertise and observe learner behaviour in order to build a learner model. In this paper we outline the process of structuring technical text educational material based on text comprehension theory in the web-based environment of ReTuDiS (Reflective Tutorial Dialogue System), using its’ authoring tool, ReTuDiSAuth with the aim to enhance web document accessibility. The technical text presented as an example concerns “Local Network Operation”. In this work we also report on evaluation results of the use of ReTuDiSAuth as an authoring tool which supports web document accessibility.

2 Text Comprehension Theory Ιn order to examine the representation constructed by learners during the comprehension process of a text, primary role should be attributed to the understanding of the cognitive categories: entity, state, event and action [3]. The term entity refers to the atoms, units or persons participating in the representation structure. The term state describes a situation in which no change occurs in the course of time. The term event refers to an effect, which causes changes but is not provoked by human intervention. The event can be coincidental or provoked by non human intervention, e.g. by a machine. An action causes changes but is originating by a man. Text comprehension is considered as the attribution of meaning to causal relations between occurrences in the text. Learners construct a representation of the text, which contains the cognitive categories: entity, state, event and action. For the interpretation of learners’ cognitive processes their discourse is analysed, in order to trace the recognition (or not) of the cognitive categories.

Enhancing Web Document Accessibility by Authoring Texts

599

Furthermore, text analysis in relation to the cognitive categories does not suffice [3]. The organization and structure of cognitive representation should involve three system types: relational system, transformational system and teleological system. The relational system represents a state in which there are entities of the possible world and no change occurs in the course of time, whereas part/ all relations (for example, among entities) define a hierarchy in the structure of the system. The transformational system represents complex events of the world or events’ sequences which provoke transformation of static states. When a transformational system is causal then it is described as a causal path among events. When it is temporal the changes are temporal. Part/ all relations between events and macro- events define a hierarchy in the system. The teleological system is organized in a tree of goals and sub-goals and within a time period its’ initial state, defined by the present entities, their relations and the values of their properties, changes turning into a final state performing in that way the predefined goal. The organization and structure of cognitive representation should also be examined on micro and macro-levels. On a micro-level scale, the creation of a text that allows a precise description of a technical system and facilitates readers in constructing its microstructure representation must involve: (a) the description of the units that constitute the system based on the causal relations which unite them and (b) the description of event sequence taking place in these units in respect to causes affecting them as well as to changes that bring the system from one state to another.On macro-level, the development of the macrostructure by readers is achieved through the reconstruction of the microstructure and the establishment of a hierarchical structure with goals and sub-goals. The creation of a text which facilitates readers in constructing its macrostructure representation for a system must involve the teleological hierarchical structure of goals and sub-goals of the various operations as well as their implications.

3 Web Authoring Tools ITS authoring is both a design process and a process of knowledge articulation. While authoring tools are becoming more common and proving to be increasingly effective they are difficult and expensive to build. Web authoring promotes moreover universal accessibility to authoring tools. Universal accessibility is integrated in the user interface development life-cycle. Authoring tools use methods to achieve the following goals: [1] a) decrease the effort of authoring (time and cost), b) allow others to take part in the design process c) help the author articulate or organize his domain knowledge d) support good design principles concerning the pedagogy and the interface and e) allow quick evaluation cycles. Authoring tools achieve the above goals using a number of methods in order to simplify and automate authoring and knowledge acquisition. Part of ITS authoring is the systematic decomposition of the subject matter into a set of related elements, for example a hierarchy. Each authoring system provides tools or cues which assist the author in this process of breaking down and elaborating the content to the necessary level of detail according to an instructional model. Authoring tools allow non-programmers to build tutors by incorporating a particular model or framework to scaffold the task [15]. Learner modelling process requires

600

G. Tsaganou et al.

making certain choices, and it is in these choices that the learning process is located. We do not learn much from looking at a model, we learn from models by building them and using them [11]. Learning from building models involves finding out what elements fit together in order to represent the world of the model. The design of dialogue activities for adaptive learning supported by appropriate authoring tools attracts the interest of many researchers and educators in inventing new methodologies for effective teaching and learning. The authoring process activates authors to decompose the subject matter into a set of related elements to discover what elements fit together in order to represent a concept, for example a hierarchy. The authoring process, as a process of choosing, organizing, structuring and linking educational material becomes a process of learning. Authoring tools for text comprehension have to discover and offer mechanisms which help authors design activities for the diagnosis of learners’ difficulties in comprehending texts. They offer the appropriate structure and guidance in order the author to be able to import and elaborate educational material (text, questions, dialogues etc.). There has been a growing concern about scientific text comprehension [5]. Efficient teaching and learning requires that educators should be familiar with the difficulties which learners are likely to face. 3.1 ReTuDiS System ReTuDiS, consisting of two parts: diagnosis and dialogue part, is an web-based open learner modelling tutorial dialogue system for text comprehension. The system infers learners’ cognitive profile in order to construct and revise the learner model with the learners’ participation [21]. The diagnosis part of ReTuDiS approaches learner’s text comprehension supporting the theory of Baudet & Denhière that learner’s representation of the text contains the cognitive categories [3]. The system engages learners in an activity which includes reading comprehension of text and text activities. The diagnosis part infers learners’ cognitive profile and his learner model. The underlying theory beyond the dialogue part of ReTuDiS is the Theory of Inquiry Teaching [6]. ReTuDiS approaches dialogue activities based on theories of dialogue management, strategies, tactics and plans which promote reflection in learning. The dialogue part is based on the learners’ cognitive profile, inferred by the diagnosis part, the learners’ answers to text activities and the selected dialogue strategy offered by the system. The dialogue part of ReTuDiS engages the learner in personalized reflective dialogues in order to revise the learner model with the participation of the learner. 3.2 ReTuDiSAuth ReTuDiSAuth, the web authoring tool of system ReTuDiS, offers an environment that supports users registered as authors or administrators [9], [21]. Authors have the authority to add new texts, to restructure them and create activities or edit existing ones. Administrators of the system have the authority to manage the base of users and the educational material. Guidance, offered by the system during authoring, provides universal accessibility for authors. The environment offers the following tools (Figure 1):


601

Fig. 1. The ReTuDiSAuth: Screenshot of the tool for structuring text

Text fields. These fields are designed to help authors enhance their own educational material into the system for example, texts, titles of activities, text activities. Pop-up menus. The menus are designed to help authors select from predefined by the author or by the system values or defaults such as: categories of activities, characterizations of answers. The knowledge base. This base includes the educational material of text, questions, activities etc. Association buttons/semantic relations. They are buttons designed to help authors establish temporal, causal or part /all relations among text elements and make associations between elements of educational material semantically related. Guidance tips. They are information tips which contain accessibility guidelines to support the author by giving back the appropriate feedback to his actions for example, confirmation or not of the completion of each step. Administrative tools. These tools are for managing the lists of users, the roles of the users (teachers or students), the categories of activities, the activities themselves, and some reports on carried out activities (log files for each student).

4 Structural Analysis of Text School and university text-books usually include texts structured not according to any theory of text comprehension. Authors of such texts usually ignore micro and macro structure. Research held with participation of 60 students studying Didactics of

602

G. Tsaganou et al.

Informatics in the Department of Informatics and Telecommunications, University of Athens during the academic year 2006-2007, asked students to select texts and write questions for text comprehension [10], [20]. The research results indicated that selected texts embody mainly descriptions of micro structure, whereas descriptions of macro structure were very poor or fragmentary. On the other hand, questions reported by the students included descriptions of macro structure. Structuring a text is a demanding process. The text should be organized and structured in order to include descriptions on micro and macro-level representation. Since this is difficult, authors lie heavy on the construction of the appropriate text activities about the text. ReTuDiSAuth involves the author in the following processes concerning text and text activities. 4.1 Authoring Text Structure The author 1) selects the text, which describes a technical system, 2) identifies the cognitive categories: entities, events, states, actions as well as causal, temporal and part/ all relations and 3) restructures the text so as to include: (a) on micro-level descriptions of entities, part/ all relations of entities, states, events and event sequences, which provoke changes to the state of the system, temporal and causal relations and part /all relations among events and macro-events and (b) on macro-level descriptions of goals and sub-goals and how the technical system changes from an initial to a final state due to events in order to achieve the goals and sub-goals. 4.2 Authoring Text Activities Each text activity refers to a description on micro-level or macro-level. According to the relational system, it refers to: (a) a description of units that constitute the system, (b) a description of part /all relations connecting system units and (c) a description of static states of the units. As far as the transformational system is concerned: (a) a description of events and events’ sequence taking place in these units, (b) a description of causal and temporal relations among events and the changes they bring to the state of the system. Finally, with respect to the teleological system a description of the system on macro-level throughout a “tree” of predefined goals and sub-goals for every transformation of the system from one state to another is included. We propose the following activities: a) Categorizing entities To structure the text activity the author, guided by the authoring tool, inserts in the appropriate fields, entities described in the text, which have part/ all relations among them and makes the appropriate connections among entities so as to declare their part/ all relations. The activity asks the student to categorize entities. The aim of this activity is student to be able to identify entities, to recognize part/ all relations among them and to categorize entities. b) Classifying events or operations To structure the text activity the author inserts in the appropriate fields one by one the events belonging to a sequence so as one or more of them appear in a wrong causal and temporal order. This activity asks the students to put events in the right order taking into account causal and temporal relations among them. The aim of this


603

activity is student to be able to identify events and sequences of events, to identify the initial and the final state of the system provoked by the events, to recognize causal and temporal relations among events, and put them in the right order the system to be able to achieve a goal or a sub-goal. c) Activities of completion of event or operations missing in a sequence To structure the text activity the author abstracts one or more of the events in a sequence of events, which sequence constitutes an operation. Student can select missing events from a list. The aim of this activity is to make student able to complete events missing in a sequence, which show temporal, causal and part/ all relations. The following technical text comes from a university textbook [4]. Text Example: Local Network Operation A computer network is often classified as being either a local area network (LAN), a metropolitan area network (MAN) or a wide area network (WAN). Another means of classifying networks is based on the topology of the network, which refers to the pattern in which the machines are connected. Three popular topologies are: (1) bus, in which the machines are connected to a common communication line called a bus, (2) ring, in which the machines are connected in circular fashion and (3) star, in which one machine serves as a central focal point to which all the others are connected. A bus topology is designed with each node connected directly to a highdata speed bus. All devices are connected to a central cable, called the bus or backbone. Nodes communicate across the network by passing packets of data through the bus (they read and write data -in the form of packets). Packets placed on the bus, transfer messages to nodes. A message includes the receiver’s address, which specifies the network address of the target node. A node watches the bus continuously and reads the target address of each packet. After that, the node compares the address with its own, and if they are the same, then reads the message of the packet, otherwise ignores it. When a node is ready to broadcast a message, waits until the bus is free and then begins passing it to the bus. If a node uses the bus it watches it and can be aware of any other node using the bus at the same time. In that case both nodes stop using the bus waiting until one of them accidentally attempts to us it. When a limited number of packets are simultaneously transmitted throughout the bus, them this competence strategy is successful. The bus topology network can work even in case of disconnection of a node. Bus network topology uses a broadcast channel which means that all attached nodes can hear every transmission and all nodes have equal priority in using the network to transmit data. A bus topology connects each computer (nodes) to a single segment trunk (a communication line, typically coax cable, which is referred to as the 'bus'. The signal travels from one end of the bus to the other. A terminator is required at each to absorb the signal so as it does not reflect back across the bus. A media access method called CSMA/MA is used to handle the collision that

604

G. Tsaganou et al.

occurs when two signals placed on the wire at the same time. The bus topology is passive. In other words, the computers on the bus simply 'listen' for a signal; they are not responsible for moving the signal along. Bus networks (not to be confused with the system bus of a computer) use a common backbone to connect all devices. A single cable, the backbone, functions as a shared communication medium that devices attach or tap into with an interface connector. A device wanting to communicate with another device on the network sends a broadcast message onto the wire that all other devices see, but only the intended recipient actually accepts and processes the message. Activity 1: Categorizing entities Title: Identification of local network units Description: Read the text carefully. You have at your disposal a list of entities and a list of categories concerning Local Network Operation. a) entities: server, bus, terminator, cable, connector, wire, backbone and b) categories: nodes, communication lines, connection devices. Identify entities, recognize part/ all relations among them and categorize entities into one from the above categories.

5 Evaluation Formative evaluation, concerning the use of ReTuDiSAuth for text restructuring and for construction of text activities aimed at evaluating web document accessibility and at further revisions, modifications and improvements of ReTuDiSAuth as well as of ReTuDiS system. It was conducted with the participation of 26 postgraduate students and 6 experts in informatics domain at the University of Athens. Students were asked to participate in the evaluation process and perform representative tasks: (a) analyze texts according to text comprehension theory, (b) restructure text and (c) prepare text activities. Experts made data collection and data analysis of the material proposed by the students. Both students and experts were given an evaluation questionnaire concerning universal accessibility of documents, usability, learnability and efficiency of the authoring tool asking: (a) how easy non-programmers can learn to use the system, (b) the amount of resources needed and how quickly a trained user can restructure texts and construct text activities. Experts commended about the quantity and the quality of texts restructured and text activities constructed by the students. They identified as beneficial the method used for enhancing accessibility of documents and for the improvement of students’ authoring skills for web-based text activities.

6 Conclusions This work contributes towards universal accessibility of documents for students and authors. ReTuDiS environment offer students access to technical texts and text activities. ReTuDiSAuth offer authors an authoring tool and guidance to restructure texts and create text activities for students. Research results concerning accessibility and effectiveness of the environment as an authoring tool for structuring educational text material were presented.


605

Currently, we are exploring improvement of the system concerning text structuring for personalization. The system, after a pretest concerning student’s prior knowledge offers three texts in respect of the relational, the transformational and the teleological system. Moreover, as ReTuDiS does offer significant advantages for universal accessibility throughout the web can be tested in real classroom conditions.

References 1. Ainsworth, S., Major, N., Grimshaw, S., Hays, M., Underwood, J., Williams, B., et al.: R.: REDEEM: Simple Intelligent Tutoring Systems from Usable Tools. In: Murray, T., Blessing, S., Ainsworth, S. (eds.) Authoring Tools for Advanced Technology Learning Environments, pp. 205–232. Kluwer Academic Publishers, The Netherlands (2003) 2. Arruarte, A., Ferrero, B., Fernandez-Castro, I., Urretavizcya, M., Alvarez, A., Greer, J.: The IRIS Authoring Tool. In: Murray, T., Blessing, S., Ainsworth, S. (eds.) Authoring Tools for Advanced Technology Learning Environments, pp. 233–267. Kluwer Academic Publishers, The Netherlands (2003) 3. Baudet, S., Denhière, G.: Lecture Comprehension de Texte et Science Cognitive. Presses Universitaires de France, Paris (1992) 4. Brookshear, J.G.: Computer Science: An Overview, Pearson International Edition, 9th edn. (2005) 5. Brown, A.L., Day, J.D.: Macrorules for summarizing texts: The development of expertise. Journal of Verbal Learning and Verbal Behavior 22, 1–14 (1983) 6. Collins, A.: A Sample Dialogue Based on a Theory of Inquiry Teaching. In: Reigeluth, C. (ed.) Instructional Theories in Action, pp. 181–199. Lawrence Erlbaum Associates Inc., Hillsdale (1987) 7. DeCorte, E., Verschaffel, L., DeWin, L.: Influence of rewording verbal problems on children’s problem representations and solutions. Journal of Educational Psychology 77, 460– 470 (1982) 8. Graesser, A., Tipping, P.: Understanding Texts. In: Bechtel, W., Graham, G. (eds.) A Companion to Cognitive Science. Blackwell, Malden (1999) 9. Grigoriadou, M., Tsaganou, G., Cavoura, T.: Historical Text Comprehension Reflective Tutorial Dialogue System. Educational Technology & Society Journal, Special issue 8(40), 31–40 (2005) 10. Grigoriadou, M., Tsaganou, G.: Authoring Tools for Structuring Text Based Activities. In: Proceedings of the 4th International Conference on Universal Access in Human-Computer Interaction (UAHCI 2007), Beijing, P.R. China, vol. 7, pp. 319–328 (2007) 11. Jonassen, D.: Model Building for Conceptual Change: Using Computers as Cognitive Tools. In: Grigoriadou, M., Raptis, A., Vosniadou, S., Kynigos, X. (eds.) Proceedings of the 4rd Panellenic Conference with International Participation: Information and Communication Technologies in Education (ETPE2004), Athens, Greece, pp. 4–17 (2004) 12. Kintsch, W.: Compréhension: a paradigm for cognition. Cambridge University Press, UK (1998) 13. Koedinger, K., Anderson, J.: Intelligent Tutoring Goes to the Big City. International Journal of Artificial Inteligence in Education 8, 30–43 (1995) 14. McNamara, D.S., Kintsch, E., Songer, N.B., Kintsch, W.: Are good texts always better? Text coherence, background knowledge, and levels of understanding in learning from text. Cognition and Instruction 14, 1–43 (1996)

606

G. Tsaganou et al.

15. Muray, T.: An Overview of Intelligent Tutoring System Authoring Tools: Updated analysis of the state of the art. In: Murray, T., Blessing, S., Ainsworth, S. (eds.) Authoring Tools for Advanced Technology Learning Environments, pp. 491–544. Kluwer Academic Publishers, The Netherlands (2003) 16. Ritter, S., Blessing, S.: Authoring tools for Component-Based Learning Environments. Journal of the Learning Science 7(1), 107–132 (1998) 17. Schultz, K., Bratt, E.O., Clark, B., Peters, S., Ponbarry, H., Treeratpituk, P.: A Scalable, Reusable, Conversational Tutor: SCoT. In: Proceedings of the 11th International Conference on Artificial Intelligence in Education Workshop: Tutorial Dialogue Systems, Sydney, Australia, pp. 367–377 (2003) 18. Klironomos, I., Antona, M., Basdekis, I., Stephanidis, C.: White paper: promoting Design for all and e- accessibility in Europe. Universal Access in the Information Society 5(1), 105–119 (2006) 19. Tsaganou, G., Grigoriadou, M., Cavoura, T., Koutra, D.: Evaluating an Intelligent Diagnosis System of Historical Text Comprehension. Expert Systems with Applications 25(4), 493–502 (2003) 20. Tsaganou, G., Grigoriadou, M.: Text and Dialogue Structure Analysis Matching a Theory of Text Comprehension. In: Komis, B. (ed.) Proceedings of 4th Hellenic Conference Didactics of informatics, Patra, Greece, pp. 333–342 (2008a) 21. Tsaganou, G., Grigoriadou, M.: Design of Text Comprehension Activities with RETUDISAuth. In: Pinder, S. (ed.) Advances in Human-Computer Interaction, pp. 161–172. ITech Education and Publishing, Vienna (2008b) 22. Van Dijk, T.A., Kintsch, W.: Strategies of discourse comprehension. Academic Press, New York (1983)