ICCIR'2013 International Conference

This research investigates the current software usability practices in developing countries, particularly in Uganda. It documents the contemporary user needs, attitudes and challenges that are being experienced in regard to software usage and adoption. The performance of software in circulation is judged based upon indicators like efficiency, effectiveness and ease of use. In addition, local and internationally produced software are compared in terms of functional performance and user satisfaction. This study proposes some recommendations about how to further improve software usability within the present circumstances, as well as some predictions about the future trends.

This paper considers the scheduling of n independent jobs on m non-identical machines using the ideas from game theory and multi-agent systems. The values of n and m are fixed at 30 and 6 respectively giving a schedule space with a dimension of approximately 1023 schedules. The agents are used to represent the jobs and they select machines on which the jobs should be processed, resulting into schedules. The schedules that are generated are evaluated using the makespan which is the total time taken for all the jobs to be processed. The makespan of the schedules that are generated vary when the agents that represent the jobs change the way they make their selection decisions. The agent selection policies that are investigated in this paper include pure random choice, potential game strategy and dispersion game strategy. The results that are obtained show that the random choice strategy and the potential game strategy generate the empirical best schedules by chance. The dispersion game strategy however is shown to converge very quickly to a stable schedule type whose best makespan value is between 3.1 to 3.4 times larger than the empirical best schedule. The main contributions in this paper include generating schedules in a concrete schedule space using ideas from game theory and multi-agent systems and the results that are obtained.

Resource allocation involves assigning the resources to the resource users. Where the resources and the resource users do not change with time, the resource allocation problem can be solved as a scheduling problem. Where the resources and the resource users change with time then some different allocation mechanisms are needed. In this paper an environment is considered in which the resources emerge and the resource requests also emerge unpredictably as time goes. The resources emerge with different capacities and in the same way the resource requests also emerge from the users with different demands as time goes. In this dynamic environment the resource allocation performance of the first price sealed bid and the Vickrey auctions are explored and compared. The system allocation performance is measured using the number of the emerging resource provisions and the resource requests that get matched. The simulation results show that there is no performance difference between the two mechanisms apart from the fact that the Auctioneer’s earnings are higher for the first price sealed bid auction.

In recent years, research in Machine Translation has greatly benefited from the increasing availability of parallel corpora. Processing the same text in two different languages yields useful information on how words and phrases are translated from a source language into a target language. To investigate this, a parallel corpus is typically aligned by linking linguistic tokens in the source language to the corresponding units in the target language. An aligned parallel corpus therefore facilitates the automatic development of a machine translation system. In this paper, we describe data collection and annotation efforts and preliminary experiments with a parallel corpus English - Kiswahili.

Named entity recognition is a preprocessing tool to many natural language processing tasks, such as text summarization, speech translation, and document categorization. Many systems for named entity recognition have been developed over the past years with substantial success save for the problem of being domain specific and making it difficult to use the different systems across domains. This work attempts to surmount the problem by proposing the use of domain independent features with a maximum entropy model and a multiobjective genetic algorithm (MOGA) to select the best features. The methods used in this work are backed up by experiments of which the classifications are evaluated using two diverse domains. Conclusions are finally drawn and the outlook for future work is considered.

Natural language access to databases is a research area shrouded by many unresolved issues. This paper presents a methodology of comprehending Swahili NL statements with an aim of forming corresponding SQL statements. It presents a Swahili grammar based information extraction approach which is thought of being generic enough to cover many Bantu languages. The proposed methodology uses overlapping layers which integrate lexical semantics and syntactic knowledge. The framework under which the proposed model works is also presented. Evaluation was done through simulation using field data on corresponding flowcharts. The results show a methodology that is promising.

We formulate a general Association rule mining model for extracting useful information from very large databases. An interactive Association rule mining system is designed using a combination of genetic algorithms and a modified a-priori based algorithm. The association rule mining problem is modeled as a multi-objective combinatorial problem which is solved using genetic algorithms. The combination of genetic algorithms with a-priori query optimization make association rule mining yield fast results. In this paper we use the same combination to extend it to a much more general context allowing efficient mining of very large databases for many different kinds of patterns. Given a large database of transactions, where each transaction consists of a set of items, and a taxonomy (is-a hierarchy) on the items, we find associations between items at any level of the taxonomy. We show how the idea can be used either in a general purpose mining system or in a next generation of conventional query optimizers.

Collaborative filtering is a method of making predictions about the interests of a user based on interest similarity to other users and consequently recommending the predicted items. There is a widespread use of collaborative filtering systems in commercial websites, such as Amazon.com, which has popularized item-based methods. There are also many music and video sites such as iLike and Everyone’s a Critic (EaC) that implement collaborative filtering systems. This trend is growing in product-based sites. This paper discusses the implementation of graph-based semisupervised learning methods and memory-based methods to the collaborative filtering scenario and compares these methods to baseline methods such as techniques based on weighted average. This work compares the predictive accuracy of these methods on the MovieLens data set. The metrics used for evaluation measure the accuracy of generated predictions based on already known, held-out ratings that constitute the test set. Preliminary results indicate that graph-based semi-supervised learning methods perform better than baseline methods. However, some of the memory-based methods outperform the graph-based semi-supervised learning methods as well as the baseline methods.

The literature on business process reengineering (BPR) has promoted business process reengineering as an essential element in helping organizations to achieve their goals. The literature has further claimed that Information Technology (IT) is a major enabler of business process reengineering. Despite all these claims, there has been hardly any effort to demonstrate the interrelationship between business process and information technology infrastructure, and which modeling technique can be used to demonstrate the relationship. This paper investigates how simulation modeling can demonstrate the impact of IT infrastructure on business process. The dynamic synthesis methodology is used with data collected from five commercial banks. The same collected data is used to validate the model. The results indicate that network infrastructure and data integration were significantly related to the process improvement thrust, such that network infrastructure is critical in improving business processes and enhancing customer service.

There has been an emergence of various ontologies describing data from either clinical or biological domains. Associated with this has been the development of integration systems, attempting to combine such biomedical ontologies using various strategies to overcome issues of scope and differing levels of granularity. However, such ontology based integration systems still find little use in distributed computing environments. This is attributed to both the lack of knowledge about user needs for such systems and the absence of a general framework to assess their relative suitability for specific applications. This paper aims to bridge this gap by proposing such a reference framework. The framework draws on existing information systems and ontology evaluation approaches in relating user objectives to ontology characteristics. The paper suggests that such a framework bridges the gap by emphasizing the dynamics of a biomedical environment. The framework therefore includes feedbacks from the evaluation process to the user characteristics of the integrated systems. This reference framework was validated in a study using structured interviews and a survey. The results indicate that the framework is sufficiently flexible for evaluating ontology based biomedical integrated systems, taking into account the conflicting needs of different users interested in accessing complex libraries of biomedical data.

Increasingly, information and communication technologies (ICTs) are being employed to share, generate and analyse information. In response to their decreasing cost, an increase in accessibility, and an increase in the adoption of modern ICTs across the globe, non-government organizations (NGOs) in developing countries are not only welcoming these technologies into their communities, but are using them to connect to resources outside their communities. On the lowest rung of a country’s organizational ladder, are grass-root NGOs; these small organizations are often the life line for isolated communities, and it is through them, developing countries are making progress – both economically and socially, towards advancing a country’s development. This paper presents a pilot project that explored the adoption of information technology by a small grass-root NGO in Uganda. We found the adoption of an easy to use information system together with skills training in the use of the information system and communication technologies opened up opportunities, pride and professionalism in the NGO, with a flow on effect to the people within the communities they serve. The paper concludes with several guidelines that could be followed when facilitating the adoption of ICT by organisations in developing countries.

Increasingly, information and communication technologies (ICTs) are being employed to share, generate and analyse information. In response to their decreasing cost, an increase in accessibility, and an increase in the adoption of modern ICTs across the globe, non-government organizations (NGOs) in developing countries are not only welcoming these technologies into their communities, but are using them to connect to resources outside their communities. On the lowest rung of a country’s organizational ladder, are grass-root NGOs; these small organizations are often the life line for isolated communities, and it is through them, developing countries are making progress – both economically and socially, towards advancing a country’s development. This paper presents a pilot project that explored the adoption of information technology by a small grass-root NGO in Uganda. We found the adoption of an easy to use information system together with skills training in the use of the information system and communication technologies opened up opportunities, pride and professionalism in the NGO, with a flow on effect to the people within the communities they serve. The paper concludes with several guidelines that could be followed when facilitating the adoption of ICT by organisations in developing countries.

This paper explores on how the blended learning method has been implemented in the developing country such as Uganda. Many higher education institutions in Uganda have acquired a new form of learning using technology that they combine with the traditional learning methods. Implementation of this form of learning using technology has been challenged by several factors. Many of the challenges are attributed to the conditions that exist in a developing world such as lack of appropriate technology. In spite of the several challenges blended learning has been implemented in relation to the existing internal procedures to suit the developing world context. The study strategy aimed to identify if the procedures initiated for implementing blended learning in the developing world were suitable. A case study approach was adopted in order to inspect these implementation procedures used within a higher education setting. The results from the study indicate that despite the lack of the appropriate technologies, infrastructure, pedagogical knowledge, quality assurance procedures and capital to invest, hybrid policies/frameworks are implemented. For this reason, a lot of higher education institutions are falling short of the implementation policies adopted. Therefore the paper stipulates that it is not simply adopting what the developed world has done but designing what suits the contextual setting that the particular country posses.

In this paper it has been suggested that the findings of research on managerial decisionmaking may be relevant in IT, but that research needs to be carried out on the factors that influence the decisions of IT managers in order to develop an effective “flight Simulator”. While this is useful as a first step, the paper argues that a systems thinking/ systems dynamics approach is necessary for developing flight simulators as learning environments for managers. The advantage of SD is not merely that it captures the complexity of decision-making processes; it also offers a way of exploring the impact of change on decision-making over time. The paper describes the model, identifies a number of propositions that can be derived from it and suggests how these may be tested empirically through a flight simulator.

The paper develops a system dynamics simulation model to understand the dynamics of immunisation with the aim of aiding decision making process by proposing policies that would enhance immunisation utilization. The model is designed with the intent to show how particular variables influence immunisation demand and coverage rather than predict immunisation coverage rates. The paper applies system dynamics modeling and field study research methods to capture the complex and dynamic nature of the immunization process, to enhance the understanding of the immunization health care problems and to generate insights that may increase the immunization coverage effectiveness. The model is divided into four sectors interacting with one another. The paper suggests key leverage points which could substantially improve immunisation demand, effectiveness of the health system as well as vaccine management.

Geospatial information systems provide avenues to fulfill users’ quest for both absolute and relative locations of features/objects. To achieve this, the users need to have access to different geospatial data sets from various sources. This calls for integration of data from different geospatial sources. But, there are shortcomings as different data sets may not map exactly onto one another, one of reasons being the difference in features’ geometry. This paper presents a conceptual model for geospatial data integration that can identify and measure differences and adjust spatial geometries of geospatial features to form meaningful objects which can be used for geo-spatial analysis, modeling, and easy geo-information management.

Research has shown that it is through clustering and interorganizational relationships that SMEs can be innovative, therefore creating development opportunities. Literature stresses that the success of these clusters and inter-organizational systems can be traced to the system of social and economic relationships based on trust among other factors. It is therefore a great concern for researchers to define processes that would enhance and contribute to trust. On the other hand research has shown that ICT and networked computing provide geographical clusters and interorganizational systems with new development opportunities, drawing new possible trajectories of evolution. One specific area where ICT provides new development opportunities is by reinforcing intra-cluster and inter-organizational relationships. Many social scientists have been skeptical about the contribution of ICT towards trust arguing that trust is only built where people have face-face communication, have demographic similarity, have prior history of relationship and are co-located. However it has been noted that Interorganizational systems exist that are enabled by ICT, they are located at different physical locations and have no prior relationships. These in itself creates a gap for researchers to be able to explain the existence of such relationships. By considering different dimensions of trust this paper posits that ICT contributes to creation and sustenance of trust, which is a key component for the success of inter-organizational systems. The paper seeks to provide a framework upon which ICT is analytically evaluated for its contribution towards creation and sustenance of trust.

The loss of twenty five million (25) child-benefit records was disclosed on 20th November 2007 to the House of Commons. The scale of the security lapse at the Treasury’s tax agency, affected almost half the British population. Two computer discs containing the names, addresses and dates of birth of nine and a half (9.5) million parents and fifteen and a half (15.5) million children, together with their national-insurance and child-benefit numbers and bank-account details, went missing. The records would be a treasure trove for identity fraudsters. Such large scale institutional data loss is not an isolated incident. Whereas personal data loss by or stolen from credit-card companies, online retailers, government departments and banks is well known, there now appears to be no point in fraudsters stealing one person’s details when records can be had in their thousands and millions from processes and computer systems that are not secure. Moreover such theft is not limited to public sector or the UK. The policy implications and regulatory issues and possible sanctions are discussed.

In this paper we analyze the potential for ICT enabled services by Ugandan firms to both clients within Uganda and outside Uganda. We highlight the opportunities and challenges of providing ICT-enabled services by Ugandan firms to companies, organizations, and government departments within Uganda and outside Uganda. We also suggest ways of addressing a common challenge of managing and meeting client expectations in ICT-enabled sourcing services and highlight key recommendations.Categories and Subject Descriptors: J.1 [Computer Applications]: Administrative Data Processing-Business, education, marketing; K.4.2. [Computers and Society]— Social Issues—employment; K.4.3. [Computers and Society]—Organizational Impacts—employment.

In this paper we describe the localisation of iLanga, a VoIP PBX system, into isiXhosa. In South Africa, for social and political reasons, making systems available in the African languages is becoming more and more important. On the one hand, it gives access to the use of technology to people who are not fluent in English. On the other hand, it has a strong symbolic value. Although a variety of software is already available in a number of African languages, localisation in such language is still in its early stage. The process presented interesting challenges both from the technical and the linguistic point of view.

A lot has been done in the use of Information and Communication Technology (ICT) in business yet very little in solving agricultural problems in developing countries. The problem addressed in this paper is the integration of widely available knowledge, processing and providing effective delivery for countries with low ICT infrastructure. Data was collected from small-scale farmers, meteorologists, agricultural extension officers and a research station and analyzed. From this, a suitable knowledge based model was developed. In order to evaluate the model, a prototype was developed and used in tests with sample data. The results reveal that the best way of disseminating agro-meteorological information to small scale farmers is using an integrated knowledge management system and mobile phone technology.

Whereas there are diverse university funding sources including: funding/research grants from government, research contracts from private sector and research councils, tuition fees from students, and internally generated funds through short courses and consultancies, it is widely acknowledged that universities worldwide are facing growing difficulties in financing their activities. It is further claimed that in the times of financial constraints in universities, teaching is favourably emphasised, rendering research and development deficient. This paper explores the impact of funded research projects on perceived quality of university using systems thinking and system dynamics simulation modelling. A conceptual system dynamics model based on current literature and supported by research practice in the Ugandan universities is used to demonstrate the effects of research funding on: new research resources, staff development, quality of graduate outcomes, and demand for graduate training. The authors envisage that the model developed may provide the basis of testing various acknowledged theories on the problem of funding university research projects.

The liberalization of education has fostered an unprecedented demand especially at the tertiary level. In several deliberations, human capital ultimately emerges as the decisive factor governing the success of any endeavor of scale and substance. In the Information and Communications Technology (ICT) era, as indeed in any other era, the needs of all people cannot be completely fulfilled. Even now the power of ICT has been fully harnessed to apply to a relatively small portion of humanity even though nations, institutions, and individuals are continually thinking of how to use this technology to reach the marginalised. The principal difficulties include the galvanizing of people and aligning different cultures to forge ahead with initiatives that converge on endeavors to satisfy human aspirations, be they physical, emotional, mental, or even spiritual. The development of human capital with the requisite skills, understanding, motivation, vision, focus and commitment without vexing restrictions or constraints are essential attributes for eventual success.The area for human resource training has been foremost in the minds of the planners. Higher funding levels, the expansion of public institutions of higher learning, and the ecognition of the role of the private tertiary education sector to enable a more effective and fast-tracked training of the requisite manpower have been most obvious steps. In the drive to achieve manpower training effectiveness, private education sector has been encouraged, resulting in timely approvals to provide university level courses, especially in capital intensive and or marketable fields like Computer science, Information Technology, Information Systems, Software Engineering, Communication, and Business Administration.This paper analyses twelve tertiary institutions (private and public) of EAC with a focus on both human resource and curriculum development and presents a snapshot of ICT education status in which the identified opportunities, risks and challenges are discussed. The paper also suggests solutions that could be considered at sub-regional level.

Due to the explosive growth of mobile and wireless technologies and services in Africa, a number of questions are being raised by academicians, technology professionals, industry and policy makers on how it is going to affect the future development of society in Africa. The recent World Wide Web consortium (W3C) and World bank organized workshop in Brazil [1,2], have discussed issues on the effects of mobile technology for social development. This was also partly motivated by the increased internet and web presence in the mobile area, and the possibility of using mobile subscriber terminals as a tool for delivering public information and services. A number of mobile projects for health, governance, banking and learning have been identified in the boundary region of mobile and web technologies. One important conclusion that came out from the discussion was that, the current few mobile SMS based services, appearing on the scene are just a proof of concept and announce the beginning of the thousands of mobile data based services that will engulf the mobile economies of the future. The main reasons for this are: 1) The ubiquity of the mobile device and the high increase in mobile subscriber base, soaring above the 3.5 billion value, and 2) The improved capabilities of mobile devices, 3) The flexibility of time and geographic independence that is possible in accessing information using mobile technologies. This paper will try to address some issues of mobile technology and sustainable economic development in a developing country context.

Wireless sensor networks (WSN) consist of a large number of small sensor nodes, usually spread out in hard accessible areas and communicating wirelessly. A sensor node Combines the abilities to sense, compute, and communicate to other nodes. The large numbers of nodes with minimum capacity devices operating in constraining and demanding real-world environments impede communication within and outside the network, making the process of implementing security in wireless networks exceptionally difficult and expensive. We discuss these difficulties and what is being done to overcome them in order to meet the ever-growing and popular wireless sensor networks. We conclude by considering several potential future directions for security solutions.

Peer-to-Peer (P2P) networks are popular in sharing multimedia resources such as video, audio and images. Its popularity has attracted very many users thus increasing data streaming along a congested path. The increased sharing of the resources in the network has caused delays making the network resource sharing inefficient. This paper presents a delay-optimized path algorithm that increases data flow within the network among the peers and utilizes the network resources for effective and efficient streaming of data. We consider a distributed search for peer nodes with multi-peer participation in both searching and initiating resource transmission within the network. The resource streaming across the peer network is controlled by the resource function. Single peer initiated searching is also considered where a single peer node initiates the resource search process. Our analysis shows that the nature of growth of the network is linear.

Future networks are characterized by convergence at different technological and service levels. Convergence of fixed and mobile network is a trend that is used to develop service oriented communication networks. This paper discusses service oriented telecom architecture for convergent networking and enablement of new innovative mobile services. Analysis of different existing migration strategies for service orientation results in a near similar IP-Multimedia Subsystem (IMS)-based architecture for the case of developing countries. The main migration and networking guidance parameters used are: capital and operational cost reduction, simplicity of introducing & managing new mobile services, unified pricing & billing for revenue generation, smooth migration of legacy network, and adherence to international telecom standards.

The paper contains results of the comparison between African (namely Rwandan) and Russian ICT educational policies. First the author presents a review of ICT educational strategy in Rwanda, including The Vision 2020 Plan, National Information and Communications Infrastructure Plans and international ICT projects in Rwanda. Then the author gives a review of ICT educational strategy in Russia, including Informatization of Education Process, Program of Computerization of Rural Schools and some non-governmental initiatives. Several parameters, like the level of ICT literacy, the degree of integration into the international educational system, technical base of ICT, the degree of syllabi unification, the syllabi contents, the degree of students’ motivation to ICT learning, have been compared. The author draws the conclusion that the Rwandan ICT educational policy being completely realized should lead to prosperity and educational superiority of the country. The Russian ICT educational policy has been quite effective for a long time but today has to be adjusted to the contemporary world.

The paper contains results of the comparison between African (namely Rwandan) and Russian ICT educational policies. First the author presents a review of ICT educational strategy in Rwanda, including The Vision 2020 Plan, National Information and Communications Infrastructure Plans and international ICT projects in Rwanda. Then the author gives a review of ICT educational strategy in Russia, including Informatization of Education Process, Program of Computerization of Rural Schools and some non-governmental initiatives. Several parameters, like the level of ICT literacy, the degree of integration into the international educational system, technical base of ICT, the degree of syllabi unification, the syllabi contents, the degree of students’ motivation to ICT learning, have been compared. The author draws the conclusion that the Rwandan ICT educational policy being completely realized should lead to prosperity and educational superiority of the country. The Russian ICT educational policy has been quite effective for a long time but today has to be adjusted to the contemporary world.

Technology has contributed a lot of improvements in pedagogical aspects of education. To many educators it is believed that technology will some day control the entire activities within education. This approach which is referred to as e-learning is transforming how education is being delivered. However the e-learning approach depends on the proper integration of technology, with the existing traditional means. The technology enabled approach should be positioned to enhance and not replace the existing traditional mean. There are a lot of ways through which technology has been adopted to enhance learning and these could include content dissemination, interaction, assessment, evaluation, communication and supervision. In this paper e-supervision is discussed as one method that technology is enhancing within higher educational institutions of learning. The paper discuses a pedagogical model for esupervision that is facilitated by the available technology. This model indicates that there are several methods that are being adopted to enhance the traditional supervision. The methods include use of e-mails, discussion boards, forums, telephony, chat rooms, wiki, blogs and e-research group. The methods can be effective in enhancing supervision but would need a strong foundation in setting up a technological infrastructure, social atmosphere, communication, solidarity, time schedule, prompt of response and respect for members.

The rapid technological development in mobile phones coupled with their rapid diffusion into all walks of life has instigated various sectors to exploit them for various purposes aimed at enhancing rganizational efficiency and flexibility. The communication, business, financial, banking and education sectors have developed and continue to develop applications for mobile phones. Our research sought to evaluate an initiative introduced at Makerere University’s Department of Distance Education (Mobile Research Supervision Initiative – MRSI), for guiding (using mobile phones) distance learners who were completing their final year field research projects. The study adopted a mixed methods research approach, employing a self administered semi-structured questionnaire/survey, in-depth interviews and document review methods for data collection. Results from the evaluation have indicated that the mobile research supervision: i) created a virtual community of practice amongst research students and their supervisors, ii) motivated lonely distance learners in the field, iii) created a customer care feeling amongst students, and iv) bred intimacy amongst the alumni and their institution. The results of this study could be used to inform the development of mobile learning policy and pedagogy for developing countries. Keywords: Distance learning, m-learning, e-learning, mobile research supervision

In order to meet the exponentially growing demand of higher education, implementation of e-learning is an ideal choice for the universities. And the introduction of the concept of blended learning model is a natural start for the implementation of elearning solutions for the universities in developing countries which is pursued by many organizations in rich economies. When organizations are in the process of implementation of new type learning, a number of factors come into play. Some factors are about the technology, others about the prospective users, still others about the local context of use and the associated costs [Wilson et al. 2002]. This paper sought to study the readiness of organizations seeking to implement blended learning in constrained low bandwidth environment with an emphasis on university settings and proposes an e-readiness model. The study employed three types of instruments; Institutional Readiness Questionnaire, Lecturers’ Perception Questionnaire, and Students’ Perception Questionnaire.

Since their inception during the first century, library institutions have used manual operations to provide services to patrons. Information seekers in these libraries spent a lot of time perusing through card catalogues and searching rows of stacks for material that may have been checked out by someone else. The tradition tool and techniques were inadequate and time consuming. The introduction of the new technologies has changed this concept and what we have now are hybrid libraries based on computer networks and physical facilities. This investigation is an effort to examine alternative measures of intention in revalidating and expanding (Venkatesh et al. 2003) UTAUT model in the context of hybrid library services using university communities in a less developed country, Uganda.

Web-based learning systems give students the freedom to determine what to study based on each individual learner’s learning goals. These systems support learners in constructing their own knowledge for solving problems at hand. However, in the absence of instructors, learners often need to be supported as they learn in ways that are tailored to suit a specific learner. Adaptive web-based learning systems fit in such situations. In order for an adaptive learning system to be able to provide learning support, it needs to build a model of each individual learner and then to use the attribute values for each learner as stored in the model to determining the kind of learning support that is suitable for each learner. Examples of such attributes are learner knowledge level, learning styles and learner errors committed by learners during learning. There are two important issues about the use of learner models. Firstly, how to initialize the attributes in the learner models and secondly, how to update the attribute values of the learner model as learners interact with the learning system. With regard to initialization of learner models, one of the approaches used is to input into a machine learning algorithm attribute values of learners who are already using the system and who are similar (hence called neighbors) to the learner whose model is being initialized. The algorithm will use these values to predict initial values for the attributes of a new learner. Similarity among learners is often expressed as the distance from one learner to another. This distance is often determined using a heterogeneous function of Euclidean and Overlap measures (HOEM). This paper reports the results of an investigation on how HOEM compares to two different variations of Value Difference Metric (VDM) combined with the Euclidean measure (HVDM) using different numbers of neighbors. An adaptive web-based learning system teaching object oriented programming was used. HOEM was found to be more accurate than the two variations of HVDM. Key words: learner modeling, initialization, web-based learning, nearest neighbors, overlap measure, knowledge level, object oriented programming