Browsing by Author "Teferi, Dereje"

Now showing 1 - 4 of 4
  • No Thumbnail Available
    Item
    The Application of Data Mining in Crime Prevention: the Case of Oromia Police Commission
    (Addis Ababa University, 2003-06) Woldu, Leul; Tadesse, Ethiopia; Teferi, Dereje
    Law enforcement agencies like that of police today are faced with large volume of data that must be processed and transformed into useful information and hence data mining can greatly improve crime analysis and aid in reducing and preventing crime. The purpose of this study is to explore the applicability of data mining technique in the efforts of crime prevention with particular emphasis to the Oromia Police Commission and to build a model that could help to extract crime patterns. With this objective decision trees and neural network were employed to classify crime records on the basis of the values of attributes crime label (CrimeLabel) and crime scene (SceneLabel). Results of the experiments have shown that decision tree has classified crime records at an accuracy rate of 94 percent when the attribute CrimeLabel is used as a basis for classification. Where as, in the same experiment, the accuracy rate of neural networks is 92.5 percent. On the other hand, in the case of classification of records on the values of the attribute SceneLabel decision tree has shown an accuracy rate of 85 percent while neural network revealed 80 percent. In both experiments the output indicated that decision tree performed better. Besides, decision tree generated understandable rules that could be easily presented in human language and thus police officers can make use of these rules for designing crime prevention strategies. Thus, this experiment has proved that data mining is valuable to support the crime prevention process and particularly, decision trees seem more appropriate for the domain problem.
  • No Thumbnail Available
    Item
    A Cloud Computing Framework for Ethiopian Higher Education Institutions
    (Addis Ababa University, 2012-06) Belachew, Sewale; Teferi, Dereje
    Educational institutions throughout the World have become highly dependent on information technology for their teaching-learning, service delivery and business requirements. Procuring and maintaining a wide range of hardware and software require substantial, ongoing investment and the skills to support them. In the current financial crisis and being challenged by growing needs, universities are facing problems in providing necessary information technology (IT) support for educational, research and development activities. The objective of this paper is to find alternatives to the use of IT, while leading universities try improve agility and obtain savings. The research methodology consisted a rigorous analysis of the latest research on Cloud Computing as an alternative to IT provision, management and security. It also took into account the best practices for Cloud Computing usage within universities, interviewing with selected ICT directors, plus the authors’ experience in IT and higher education. The research paper discussess the advanatges of cloud computing for educational institutions, the limitations of current IT utilization in Ethiopian Higher Education institutions. It also discussess alaternative solutions to solve the current IT utilizations limitations in Ethiopian Higher Education Institutions. The research finding shows that Cloud Computing is the better ICT utilization mechanism for Education institutions teaching-learning and service delivery requirements, for it enables wise and strategic use of technoloy which significantly reduces cost. Accourdingly, when the Proposed Hybrid Cloud Computing model is implemented, it will have significant contribution to the country in different aspects. Keywords: Cloud Computing, Higher Education Institutions
  • No Thumbnail Available
    Item
    Line Fitting to Amharic OCR: the Case of Postal Address
    (Addis Ababa University, 2003-07) Hailemariam, Mesay; Teferi, Dereje; Taddesse, Nigussie; Assabe, Yaregal
    Currently researchers are attracted to the area of Optical Character recognition primarily due to challenging nature of the research and secondly due to the industrial importance that it provides in the area of Reading machine for the Blind, postal Address interpretation, Bank Curtsey amount processing, hand filled form processing, and the like. Research in the area of Amharic OCR systems is ongoing since 1997. Attempts were made in adopting algorithm to Amharic language, incorporating preprocessing techniques to the adopted algorithm, and in generalizing the system so as it recognizes Type written characters as well as hand written characters. Sufficient amount of work is done in the areas of preprocessing such as segmentation and Noise Removal. However, the consideration given to the simplification of the feature extraction and the efforts made to alleviate the problems of high dimensional input still requires the contribution of many additional researches in order to come up with a system that the society can use to solve real world problems. To this end, Line fitting is used to Amharic Optical character recognition by applying simple geometric calculations to determine features which could represent and describe the character as uniquely and precisely as possible. The image of a segmented character which is normalized into 32x32 pixels is divided into 16 smaller squares of 8x8 pixels. Then the least square technique was applied to fit a linear model to the distribution of foreground pixels and three features were extracted from each smaller square. Finally, a feed forward Neural Network trained using a back propagation algorithm is used on handwriting of three individuals using a cross validation technique as well as a separate test set and results are depicted on tables and confusion matrices. Relevant Conclusions were drawn and some valid recommendations were forwarded to indicate future direction of further works on the area.
  • No Thumbnail Available
    Item
    Optical Character Recognition of Typewritten Amharic Text
    (Addis Ababa University, 1999-05) Teferi, Dereje; Fissaha, Sisay (PhD); Alemu, Worku (PhD)
    Optical Character Recognition is an area of research where a system is made to accept a document image and convert it into ASCII code so that it will be easy for storage, retrieval, and filterer processing. OCR helps to convert a bulk of information available on paper to electronically processable format without human intervention -- saving time, money, and labor Recently Optical Character Recognition for the Amharic Script has become an area of research interest. Some developments have been made in recognizing characters with specific type style, font, and font size. All the trials in this regard are on very high quality laser printouts on white papers. In reality, however, most Amharic typewritten documents that need to be converted into machine-readable format are typewritten and on non-white paper III this study an attempt is made to explore the possibilities of developing an OCR system for typewritten Amharic text. To this end, features of the typewritten Amharic characters are thoroughly studied. Some algorithms for noise removal and segmentation are reviewed. These algorithms are implemented to see their performance on typewritten Amharic text. Previous algorithm implemented for recognition of Amharic characters is modified to incorporate the specific features of typewritten Amharic characters. The segmentation and the noise removal algorithms are integrated with this algorithm. The result is tested on typewritten Amharic documents, and test results are presented. Recommendations are also drawn to point out issues to be investigated filterer for the development of typewritten Amharic OCR system with better performance.