ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION AT DELTA UNIVERSITY FOR SCIENCE AND TECHNOLOGY GAMASA, EGYPT, 4 – 6 DECEMBER 2010

PSDD : Prolog-based System for Dyslexia Diagnosis Dr. Adnan G. Abuarafah1, Dr. Osama khozium2* (Corresponding Author is Marked with a *, Presenter Author is Underlined) 1

Faculty of computer and information systems, Umm Al-Qura University, Makah, SA, [email protected] 2 Faculty of computer and information systems, Umm Al-Qura University, Makah, SA, [email protected] Abstract Dyslexia is a learning disability that can hinder a person's ability to read, write, spell, and sometimes speak. Dyslexia is the most common learning disability in children and persists throughout life. The severity of dyslexia can vary from mild to severe. The sooner dyslexia is treated, the more favorable the outcome. This paper proposes a prolog based system for dyslexia diagnosis PSDD. PSDD is not only able to determine the severity degree of this disability but it is also able to propose a treatment plan with different steps that need to be taken by both teachers and parents. Through their feedback, PSDD is able to adapt its knowledge in order to give more enhanced treatment plans.

Keywords: Dyslexia; logic programming; diagnosis; user requirements. 1. Introduction Dyslexia is not a disease. It is a condition in which a person's brain learns in a different way from that of other people. The condition has nothing to do with a person's intelligence. Dyslexia is affecting up to 17% of the population[1] [2] [3] [4] [8]. Those learning disorders are usually related to the use of language and reading. Learning disorders occur in people of all ages, races, and income levels. Children have this problem often struggle to acquire word reading and/or spelling skills and knowledge despite appropriate intervention. Figure 1 shows that people with dyslexia tend to confuse certain letters and have difficulty spelling. The British Psychological Society’s Division of Education and Child Psychology drew together widespread research showing that dyslexia has multiple causes and the appropriate intervention consists of frequently delivered and highly structured phonic learning, word reading and spelling skills programs. Their definition of dyslexia is as follows: “Dyslexia is evident when accurate and fluent word reading and/or spelling develops very incompletely or with great difficulty. This focuses on literacy learning at ‘word level’ and implies the problem is severe and persistent, despite appropriate learning opportunities.”[1] P a g e |118

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

Figure 1. A Sign of Dyslexia

An independent report, dated June, 2009, from Sir Jim Rose to the Secretary of State for Children, Schools and Families refers to dyslexia as “Dyslexia is a learning difficulty that primarily affects the skills involved in accurate and fluent word reading and spelling.”[2] Dyslexia can cause frustration and negative feelings about competencies for some children and in turn this might cause behavior problems. History gives many examples of great people who have overcome dyslexia to achieve enormous success– Albert Einstein being probably the most famous. However, Dyslexia is worrying for both parents and educators [1] [7] [9] [10]. Many teachers lack knowledge of the symptoms of dyslexia and also the resources needed to diagnose the condition. While teachers continue to carry out tremendous work in schools, they are however, under extreme pressure, which does mean some dyslexic children go unnoticed. Children therefore suffer because they do not get diagnosed early enough. In addition, current tools to diagnose dyslexia can be very expensive and many schools are unable to purchase them due to budget constraints. The opportunities for developing an intelligent system for teachers and parents are extremely valuable. An intelligent diagnostic tool can help teachers and parents identify the symptoms that are commonly displayed in dyslexic children. This expert system could allow primary school teachers to confirm that a child may need further investigation by the teachers

P a g e |119

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

themselves, the parents or by the educational child psychologist. Moreover, the system will be designed to propose a treatment plan to cure or solve the problem and symptoms of dyslexia. This paper proposes a prolog-based system for dyslexia diagnosis (PSDD) which is intended to work as a tool that is easy to use and is not expensive. As stated above, early detection is the key to helping a dyslexic child so PSDD targets both parents and teachers who are closely working to dyslexic children. PSDD will help teachers to know what type and degree of severity of their dyslexic children. Also, it will help parents to monitor the performance and progress of dyslexic children as well as his current emotional/ physical states. In addition PSDD will support both of them with recommendations to cure dyslexics, raise their level and awareness in dealing with this disorder. The rest of the paper is organized as follows. The next section provides overview of dyslexia as well as the prolog language for developing expert systems. Section three describes the main architecture of the proposed system (PSDD). Section four concludes the research.

2. Theoretical Background 2.1. Dyslexia

Children with dyslexia have difficulty in learning to read despite traditional instruction, at least average intelligence, and an adequate opportunity to learn. It is caused by impairment in the brain's ability to translate images received from the eyes or ears into understandable language. The child can become frustrated by the difficulty in learning to read, and other problems can appear that disguise dyslexia. The child may show signs of depression and low self-esteem. Dyslexia may affect several different functions. Visual dyslexia is characterized by number and letter reversals and the inability to write symbols in the correct sequence. Auditory dyslexia involves difficulty with sounds of letters or groups of letters. The sounds are perceived as jumbled or not heard correctly. Some children with dyslexia may suffer from visual stress. Visual stress is the experience of unpleasant visual symptoms when reading. Symptoms include illusions of shape, movement and color in the text, distortion of the print, loss of print clarity, and general visual irritation. Figure 2 shows simulation of one symptom of visual stress. The causes of dyslexia fall into two broad categories[1][7]. Neurological or genetic inheritance, forms of damage to the brain during or post birth. Environmental or lack of exposure to reading, reduced motivation and self-esteem in relation to reading and spelling, unvaried methods of teaching word reading and spelling, for example total use of synthetic phonics approaches will

P a g e |120

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

disadvantage children who are more successful at being able to recognize whole words but struggle to split words into sound patterns or build sounds into words. The problem for the investigator trying to pinpoint a cause is that factors within these two broad categories interrelate and neurological problems can cause environmental ones and vice versa.

Figure 2. Simulation of one Symptom of Visual Stress

Dyslexia is a difficult disorder to diagnose. If one were to take a group of dyslexic and a group of non-dyslexic adults, statistically significant differences between these groups in reading, writing and spelling would almost certainly be found. But if one took an individual adult with dyslexia, he or she may have literacy skills in the average range (although probably below that which might be reasonably expected from their intelligence and education), while an individual adult who does not have dyslexia may have below average literacy skills. There are many factors the psychologist or other health professional reviews to diagnose the disability. The testing determines the child's functional reading level and compares it to reading potential, which is evaluated by an intelligence test. The tests determine whether a child learns better by hearing information (auditory), looking at information (visual), or doing something (kinesthetic). They also assess whether a child performs better when allowed to give information (output), by saying something (oral), or by doing something with their hands (tactile-kinesthetic). The tests also evaluate how all of these sensory systems (modalities) work in conjunction with each other.

P a g e |121

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

2.2. Usage of Prolog in the Development of Expert Systems

Prolog as a logic programming language provides many possibilities to realize a task [12][13]. It is a general purpose language often associated with artificial intelligence and computational linguistics. It has a purely logical subset, called "pure Prolog", as well as a number of extralogical features. Programs written in PROLOG have behavior similar to rule-based systems written in LISP. PROLOG, however, did not immediately become a language of choice for AI programmers. In the early 1980s it was given impetus with the announcement by the Japanese that they would use a logic programming language for the Fifth Generation Computing Systems (FGCS) Project. A variety of logic-based programming languages have since arisen, and the term prolog has become generic. Iterative algorithms can be implemented by means of recursive predicates. Prolog systems typically implement a well-known optimization technique called tail call optimization (TCO) for deterministic predicates exhibiting tail recursion or, more generally, tail calls: A clause's stack frame is discarded before performing a call in a tail position. Therefore, deterministic tail-recursive predicates are executed with constant stack space, like loops in other languages. Logic programming provides many possibilities to implement a task. Solutions can be realized by applying a variety of different design strategies and programming techniques. Using constrained-based modeling (CBM) the solution space for a given programming task can be covered. The (CBM) approach has been showing great promise as a diagnostic approach which focuses on static cognitive states rather than problem solving processes. The CBM approach is proposed to model general principles of a domain as a set of constraints. A constraint is represented as an ordered air consisting of a relevance part and a satisfaction part. Where the relevance part represents circumstances under which the constraint applies, and the satisfaction part represents a condition that requires to be met for the constraint to be satisfied. A constraint is used to describe a fact, a principle or a condition which must hold for every solution. For example Figure 3 shows a principle that can be formulated as constraint.

Figure 3. An Example of a Constraint 1

P a g e |122

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

Constraints are not only used to limit facts, principles or conditions of a domain, they can also be used to specify the requirements of a task or to handle solution variations. Using the relevance part, constraints can be tailored according to an ideal solution, which represents the requirements of the given task. Ideal solutions enable to check whether the problem is answered correctly, looking at the semantics. Additional requirement, which have to be satisfied in that specific situation, can be specified in the satisfaction part. Figure 4 shows an example that specifies the constraints 2 to examine the operator required in the task.

Figure 4. An Example of a Constraint 2

If a constraint is violated, it indicates that the solution does not hold principles of a domain or it does not meet the requirements of the given task.

3. PSDD Figure 5 summarizes the main PSDD function. PSDD acquires knowledge about dyslexic

children, analyzes the acquired knowledge through its knowledge base and generates the degree of severity as well as a treatment plan to be used by both parents and educators. After applying the plan for the recommended period, parents and educators will be able to feedback PSDD with their remarks about the performance of the dyslexic children in order to get advanced plans to continue towards curing.

Figure 5. PSDD Schematic Diagram

P a g e |123

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

3.1. Architecture

The architecture of the proposed system consists of four main components. Figure 6 shows the main architecture of the proposed system and data flow. PSDD includes mainly two modules: knowledge base module and report generator module. Progressing Report

PSDD

Educators Feed back

KB Module

Dyslexic profile Report Module

Feed back Parents

Progressing Report

Figure 6 The Main Architecture of PSDD

3.2. Knowledge Acquisition

PSDD acquires knowledge from three sources: •

Test and state of the dyslexic to define the initial diagnosis;

•

Report of the case by the educators through PSDD;

•

Report of the case by the parents through PSDD.

Tests are designed to catch different aspects regarding different disabilities; they contain: •

Questions about vision, reading and spelling;

•

Questions about behavior, health, development and personality spelling;

•

Questions about disorientation;

•

Questions about writing and motor skills;

•

Questions about math and time management;

P a g e |124

PSDD : Prolog-based System for Dyslexia Diagnosis

•

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

Questions about memory and cognition.

Figure 7 shows an example in Arabic language of a state test that will be provided to PSDD.

Figure 7. PSDD Schematic Diagram 3.3. PSDD Inference

The main rule in PSDD is to calculate the result of each aspect as an individual by itself and evaluate its percentage , every aspects has two counters so if the answer is yes it adds 1 to the x counter and if maybe then it will add 1 to the s counter. See Figure 8. Examples of the facts used by PSDD are as follows: • •

Fact1: “Complains of dizziness, headaches or stomach ache while reading”. Fact2: “Confused by letters, numbers, words, sequences, or verbal explanations”.

P a g e |125

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

Figure 8. Some Calculations of the Results using CBM in Prolog

After calculating each aspect alone, the average of all aspects will be given in percentage as shown in Figure 8. PSDD gives some explanations and definitions for the aforementioned aspects to facilitate and explain the meaning for the nonprofessional persons. Figure 9 explains what it means by disorientation and the symptom for this disorder.

Figure 9. Disorientation Explanation by PSDD

Recommendations to the educators and parents are provided to facilitate the treatment of the dyslexics. See Figure 5. These recommendations are changed due to the current evaluation of the case and the progressing rate. Normally PSDD will modify the recommendations each report goes along with the progressing rate of the case. PSDD has the ability to add new recommendations or modify the old one if necessary. P a g e |126

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

Figure 10. Recommendations to Educators and Parents by PSDD

3.4. PSDD Output

Two kinds of output are generated by PSDD •

Progressing report provided by the PSDD each period of time, this period will be defined by the educators due to the progressing rate, normally this period around two to three months;

•

The progressing report will include some recommendations to the educators and the parents to guide them through the treatment of the case.

4. Discussion and Conclusion This paper discusses PSDD which diagnoses dyslexia cases in a simple way. PSDD generates periodic reports which reflect the progress of the case according to the recommendations provided to the educators and the parents. These recommendations will help them to support the dyslexics to overcome this disorder. As for future work we would like to include a sound recorder to read test questions, and to modify the fonts so it can be readable by dyslexic people, and try to include oral tests in it, such as written text and the person enters the answers as the meaning of the text.

P a g e |127

PSDD : Prolog-based System for Dyslexia Diagnosis

ICI 10 - 10TH INTERNATIONAL CONFERENCE ON INFORMATION

References [1] DECP, the British psychological society's division of education and child psychology "Dyslexia research", DECP, 1996. [2] Sir Jim Rose, “Identifying and teaching children and young people with dyslexia and literacy difficulties", June 2009. [3] Arlenc Waddell, "Identifying dyslexia”, DCSF publications, 2008. [4] Stanovich KE, “Explaining the differences between the dyslexic and the garden-variety poor reader: the phonological-core variable-difference model.” Journal of Learning Disabilities 21 (10): 590–604, 1988. [5] Warnke, Andreas " Reading and spelling disorders: Clinical features and causes " Journal European Child & Adolescent Psychiatry 8 (3): S2–S12, 1999. [6]

http://www.springerlink.com/content/m3 1740417111l8w3/?p=e21d91f12abf440186aa325a73b0c59dπ=1. Retrieved 2010-07-11.

[7] McCandliss BD, Noble KG., " The development of reading impairment: a cognitive neuroscience model ", Ment Retard Dev Disabil Res Rev 9 (3): 196–204,2003. [8] Czepita D, Lodygowska E, " Role of the organ of vision in the course of developmental dyslexia ", Klin Oczna 108 (1–3): 110–3,2006. [9] Birsh, Judith R., " Research and reading disability ", Paul H. Brookes Publishing., 2005. [10] NLM MeSH Browser., " Learning Disorders: MeSH Result ", http://www.ncbi.nlm.nih.gov/mesh/ 68004410?ordinalpos=1&itool=EntrezSystem2.PEntrez.Mesh.Mesh_ResultsPanel.Mesh_RVFull. Retrieved 2009-11-06. [11] Chris Stork, "Unification & Prolog" , http://www.ics.uci.edu, 2004. [12] Berna,p,et al. "programming techniques for prolog. Learning to build and comprehended complex information structures : prolog as a case study", Albex, 1999. [13] Berna,P., " Prolog programming, a first course ", Albex, 1990.

P a g e |128