Browsing by Author "Tadesse, Nigussie (PhD)"

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    OCR For Special Type of Handwritten Amharic Text ("Yekum Tsifet") Neural Network Approach
    (Addis Ababa University, 2004-06) Mulugeta, Wondwossen; Negi, Atul (PhD); Tadesse, Nigussie (PhD)
    Verbal and written communications, which are integral components of human society, have been tram formed by the development of the respective communication devices. Through the swift development in processing devices, the need and access to digitize printed information items by means of Optical Character Recognition (OCR) became possible. Despite the fact that most world languages are beneficiaries of this technology, the application of character recognition technology to Amharic text is a recent experience, and in its infancy stage when handwritten recognition is considered. This study is then an attempt made to develop a recognition engine for Amharic handwringer text written in a special type of writing style, which is called "Yekum Tsihuf" (የቁም ጽሁፍ)Before mechanical and electronic text processors were introduced in Ethiopia, information used to be recorded on natural materials by hand writing, animal skin being the dominant one. Those handwritten documents, wine in this writing style, hold vital information about history, tradition, religion, nature and etc., which render undeniable contribution to current and future studies. The availability of this information in an electronic form would greatly help preservation and communication. In this study, the application of handwritten character recognition with Artificial Neural Network implementation for the 231 main character set of Amharic language is all empted. The training and test data sets are produced by scribers who are trained to write text using the writing style. The study used various techniques at each phase from digitization to recognition levels. Preprocessing methods like image binarization, character segmentation, and size normalization and neural network recognition is made using Visual C++.Net and MATLAB programming environments. While segmentation rate of 95.96% is attained using stage-by-stage segmentation algorithm, recognition rate that ranges from 98.8% to 20.3% is obtained for different test cases. Based on the findings and the knowledge acquired during the experimentation, topics for filature research are also identified.
  • No Thumbnail Available
    Item
    Possible Application of Data Mining Technology in Supporting Credit Risk Assessment: the Case of Nib International Bank S.C.
    (Addis Ababa University, 204-07) Shawui, Meretework; Tadesse, Nigussie (PhD)
    Financial institutions in a nation playa crucial role in the development of its economy. The banking sector as one type offinancial institution is indisputably the new ji'ontier of economic development in a country. In this respect, banking has to be sound and safe jar its clistomers as well as jar the stability of the currency and economy of a counl1y. One factor that affects the well fimctioning of the banking sector is credit risk. This factor is also a general problem among commercial banks in Ethiopia. In order to deal with high default rates banks in other countries are making use of data mining. The possible application of data mining in the commercial banking sector of Ethiopia has also been tested by the use of neural network techflique. As credit risk is a risk type that bank managers give more emphasis in the loan disbursement process because it is one of the major reasons that cause a bank to fail, the study of the possible application of data mining needed jilrther investigation. To this end, the present study focuses on the application of data mining to support credit risk assessment taking as a case study Nib International Bank S.C.(NIB). In doing so the aim of this research was to assess the potential applicability of decision tree technique to help in the loan disbursement decisionmaking process of banks. The methodology used for this research had three basic steps. These were collecting of data, data preparation, and model building and testing. The required data was selected and extracted ji'01l/ Nib International Bank records. Then, data preparation tasks (such as data tram!ormation, deriving of new fields, and handling of missing variables) were undertaken. Decision tree data mining technique was employed to build and test models. , Several decision tree models were built and testedfor their classification accuracy and the model with encouraging results was taken to generate rules to support credit decision makers and the procedures adopted are described in this document .The peliormance of the developed model is validated using new datasets and its predictive accuracy is also tested. The result shows that the use of decision tree technique produces rules for justifiable credit decision-making and that it is the best technique that needs to be adopted for NIB bank as it presents a means of providing explanation for proposed decisions as compared to neural network techniqlles. A 1/ things considered, the existence of an electronic system to support the credit risk assessment of NIB bank will promote the services of the bank to its customers as well as minimize risk
  • No Thumbnail Available
    Item
    Possible Application of Data Minning Technology in Supporting Loan Disbusrement Activity At Dashen Bank S.C.
    (Addis Ababa University, 2001-07) Worku, Askale; Biru, Tesfaye (PhD); Tadesse, Nigussie (PhD)
    The commercial banking sector plays vital role in the economic development eflorts oFa counliy And the viability oFthe sector relies on the ability of the institutions to maintain a positive inflow of resources. And one of the factors that affect the ability of the commercial banking sectors to maintain positive flow of resources is the problem of default rates. Among the Ethiopian banking sector there is a general problem of high default rate and the commercial banks have tried to tackle this problem through different ways. One technique that has become popular in addressing problem of credit risk in other countries is data mining. Data mining technology has enabled banks in other countries to make good prediction on the probability thal a certain borrower would de/clUlt or not. BUI, so far. no commercial bank in Ethiopia has used data mining technologyfor such purposes i.e. assessment of credit risk. Thus, the objective of this research work was to see if application of data mining could also be beneficial in the Ethiopian banking context. For reasons o/Familiarity Dashen Bank S. C. was selected as a case study. The methodology employed for the research had baSically three stages. 71wse were collection of data, preparation of data and model building and testing. Data was collected Fom two kinds of documents that were available at the head office o/Dashen Bank. Then the data was prepared which included summarization, deriving of new fields and handling of missing data. The dala mining technique employed for the model building and testing was neural network. Neural network sofiware was thus used in building and testing a number of models. x From Ihe numerous frials, many models wilh encouraging resu/ls were obtained. These models indicaled Ihal dolo mining applicafionjor credil decision-making isjeasible 01 Dashen Bank. BUI one major Iimifafion was unavai/abilily oj dora in an eleclronicjorm. However, a survey in Ihe IT deparrmen.1 of Dashen. Bank suggesled Ihal Ihis problem is being duly addressed.