Introduction Formalization An Alignment Equation for Using Mind Map

An alignment equation for using mind maps to filter learning queries from Google Imran A. Zualkernan American University of Sharjah, UAE izualkernanaus.edu Mohammed A. AbuJayyab American University of Sharjah, UAE mjayyabgmail.com Yaser A. Ghanam American University of Sharjah, UAE yghanamgmail.com Abstract Search engines like Google play a critical role in life-long learning. However, the query capabilities of such engines remain simple and often yield a result set that is too large. In addition, search engines like Google rely on page ranking algorithm that represents the “collective consciousness” of millions of users. Learning about specifics often involves context. This paper shows how mind maps can be used as a contextual mechanism to specify what needs to be learned and to filter and retrieve the relevant sources of learning from the internet. In specific, a concept of “alignment” is introduced that filters only those information sources that are globally aligned to the mind map. The filter is implemented as a combinatorial optimization algorithm using Simulated Annealing.

1. Introduction

As semantic internet [1] and specialized repositories of learning objects e.g., [2] come of age to include structured information about resources on the internet, search engines like Google [3] remain the most successful search engines for life-long learning. Search engines typically work using page and link similarity to conduct a search [4]. It has been explicitly suggested that Google assigns “implicit” semantics to concepts [5]. Since page ranking is based on million of users, the implicit semantics assigned using these algorithms can be termed “social semantics.” However, how does one specify what an individual needs to learn in a specific context? One line of research addresses how effective answers to specific questions from a user can be formulated [6]. Object ontologies [7] are also being used to specify “what needs to be learned” in a particular context. Mind maps for brain-storming [8] have also been used as a part of literacy education [9]. Mind maps [10] and ontologies have also been abstracted from existing documents to discover the underlying latent semantic structure [11]. Conceptual graphs another graphical notation similar to mind maps, have also been suggested for use in “semantic queries” [12]. Finally, multiple techniques such as topological analysis, filtering, search and using a concept map to filter the results from a search engine like Google have also been proposed [13]. The research presented in this paper relies on the topology of a mind map and the response set retrieved from a search engine like Google to arrive at a filtered set of results that are “aligned” with the mind map hence providing a user-specific “contextual” search mechanism.

2. Formalization

Given a general purpose search engine e.g., Google that operates over a collection of web pages and a mind map given by a directed graph , E V G , where each V v i H represents a concept in the mind map, the goal of this research is to find a subset of web pages that are “aligned” with the mind map. More formally, a search engine returns an ordered response set k P P rs ,..., 1 representing potentially relevant pages. For a response set, j i implies that page i P is more relevant than j P within the first k search results. A query based on the mind map, returns a set of response sets } ..., , , { 3 2 1 l rs rs rs rs SRS where each response set i rs is generated by a query of the form i r v v .. š where r v is a designated root node or the primary starting concept and all subsequent nodes in the query are on a path from r v to i v in graph G. A set of response set SRS with a response set bound of k is said to be aligned with a concept Proceedings of the Sixth International Conference on Advanced Learning Technologies ICALT06 0-7695-2632-206 20.00 © 2006 IEEE map G if the following equation holds for each SRS rs i  For each G v v v k j i  , , , , , , k i j i k i j i rs rs D rs rs D v v d v v d Ÿ 1 Where d is defined to be the number of hops between i v and j v . D , on the other hand, is measured using the Levenshtein or edit distance [14] between two response sets i rs and j rs . The edit distance measures the number of changes required to change one sequence into another. Intuitively, the alignment equation 1 means that if the distance between a concept A and a concept B in the mind map is less than the distance between the concept A and another concept C, then an alignment dictates that the distance between the response set of A and response set of B should also be less than the distance between the response set of A and the response set of C. And that this should be true of all pairs of concepts in the mind map. Deriving an aligned SRS from an initial set of response sets is a combinatorial optimization problem that can be solved using Simulated Annealing [15]. Simulated annealing uses an objective function of cost or Energy E to guide the search for a solution. The Energy E for this problem is the number of violations of equation 1; for an ideal solution, the number of violations should be zero. The simulated annealing algorithm for this problem was implemented in Java and used the Google API [16] to connect and retrieve response sets from Google.

3. Evaluation