Browsing by Author "Biru, Tesfaye (PhD)"
Now showing 1 - 20 of 30
Results Per Page
Sort Options
Item Application of Collaborative Filtering Agent for Document Recommendation in Sdi System(Addis Ababa University, 2003-06) Zinab, Zehara; Biru, Tesfaye (PhD)Selective Dissemination of Information (SDI) is a personalized information filtering method used for delivering CUlTent information to users. The objective of SDI service is to reduce users' effort in their pursuit of relevant infomlation. However, the performance of SDI service based on simple matching of user interest profiles with new documents, has been inefficient. To improve the efficiency of such services agent based approaches are proposed. Collaborative filtering system is one type of agent teclmology that uses a database about user preferences to predict additional documents of items that the new user might like. It combines the opinions of humans to make personalized, accurate predictions. This study is therefore, an attempt to assess the applicability of collaborative filtering approach to SDI service. This report presents the basic ideas of collaborative filtering systems and describes the results obtained from a preliminary experiment conducted based on the data obtained from Intemational Livestock Research Institute's (ILRI) sm service. Finally, based on the experiment findings, recommendations for further research are forwarded.Item Application of Data Mining Technology to Predict Child Mortality Patterns: The Case of Butajira Rural Health Project (Brhp)(Addis Ababa University, 2002-06) Anagaw, Shegaw; Biru, Tesfaye (PhD); Teferri, Dereje (PhD)Traditionally, very simple statistical techniques are used in the analysis of epidemiological studies. The predominant technique is logistic regression, in which the effects predictors are linear. However, because of their simplicity, i.t is difficult to use these models to discover unanticipated complex relationships, i.e., non-linearities in the effect of a predictor or interactions between predictors. Specifically, as the volume qj data increases, the traditional methods will become inefficient and impractical. This in turn calls the application of new methods and tools that can help to search large quantities of epidemiological data and to discover new patterns and relationships that are hidden in the data. Recently, to address the problem of identifying useful information and knowledge to support primary healthcare prevention and control activities, health care institutions are employing the data mining approach which uses more flexible models, such as, neural networks and decision trees, to discover unanticipated features from large volumes of data stored in epidemiological databases.Particularly, in the developed world, data mining technology has enabled health care institutions to identify and search previously unknown, actionable information from large health care databases and to apply it to improve the quality and efficiency of primary health care prevention and control activities. However, to the knowledge of the researcher, no health care institution in Ethiopia has used this state of the art technology to support health care decision-making.Thus, this research work has investigated the potential applicability of data mining technology to predict the risk of child mortality based up on community-based epidemiological datasets gathered by the BRHP epidemiological study. 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 from the ten yea rs surveillance dataset of the BRHP epidemiological study. Then, data preparation tasks (such as data transformation, deriving of new fields, and handling of missing variables) were undertaken. Neural network and decision tree data mining techniques were employed to build and test the models. Models were built and tested by using a sample dataset of 1100 records of both alive and Died children.Several neural network and decision tree models were built and tested for their classification accuracy and many models with encouraging results were obtained. The two data mining methods used in this research work have proved to yield comparably sufficient results for practical use as far as misclassification rates come into consideration. However, unlike the neural network models, the results obtained by using the decision tree approach provided simple rules that can be used by nontechnical health care professionals to identify cases for which the rule is applicable.In this research work, the researcher has proved that an epidemiological database could be successfully mined to identify public health and sociology-demographic determinants (risk factors) that are associated with infant and child mortality in rural communitiesItem The Application of Machine Learning Technique (Naïve Bayes) For Automatic Text Summarization [The Case of Amharic News Texts](Addis Ababa University, 2005-06) Andargie, Teferi; Biru, Tesfaye (PhD)This study presents an approach to automatic summarization of Amharic news texts by extracting sentences in a give n document. The objective o f this study is to investigate the application of machine learning technique (naive Bayes method) to automatic summarization of Amharic news items. The focus is on how to use the naive Bayes classifier for automatic Amharic news text summarization to extra ct sentences , i. e. on how to train the na'ive Bayes to classify sentences from Amharic news texts First each sentence is represented by a set of p redefined features (attributes) (i .e. location of a sentence in a document, title words occurring in the sentence, and cue words occurring in the sentence) that Edmondson (1969) found as a good indicator in giving an optimum summary for scientific papers. In addition, the thematic words occurring in the sentence. Then the naive Bayes algorithm is used to train to classify sentences as "a summary" and "not - a summary" based on the feature vectors. For the purpose of this study 480 Amharic news articles is used . Evaluation of the result s of the experiments is done using 10-fold cross validation. Result of the experiment shows that the location feature gives the best result in the classification n o f sentences when using individual features. The results of different combinations of feature sets in which location feature is included shows better results than when location is not included. Based on the feature values estimated on the training program for the combination of all the features a prototype summarizer is developed which extracts sentences to a desired compression level.Item Application of Object-Oriented Approach for the Development of A Tax Database : A Case Study of Region 14 Finance Bureau(Addis Ababa University, 1998-05) Wolle, Mesfin; Biru, Tesfaye (PhD)The tax system of actually is the basis for government revenue by which the funds raised will provide opportunities throughout the economy for development and growth. Strong tax system, in turn, requires more than anything else timely and complete information as one of its prerequisites. The lack of information can slow down the efficiency of taxation and this in turn has impact on the process of national development. While rapid access to accurate information is critical to the tax operation, in the Ethiopian tax system, there have been reports that access to information was difficult and has been a persistent problems for a long time. Such problem has also been prevalent at Region J 4 Finance Bureau which is the case area of this study Most of the information required for tax related activities are obtained based on searches across several files of taxpayers. The fact that the information is not readily available and must be extracted from files increase the response time for action and increase the risk of error due to lost or misplacement of original documents. These circumstances pose further difficulties which deter the chasing of tax arrears. This study, therefore, was undertaken to address the tax information handling problems in the Region J 4 Finance Bureau, with the view of exploring the application of the features and capabilities provided by objects-oriented techniques in tackling the information access problems Survey instruments, consisting of questionnaire, interview as well as documentary analysis, were used to collect required data for the study. Analysis made on the basis of the data gathered revealed enough evidence to establish the weaknesses of the existing tax information handling. The analysis has also helped in pinpointing the system's requirement as well as identification of possible solution. As one alternative solution for tackling the problems in the existing manual system, development of a computer based database system is identified. In this connection, an attempt is made to demonstrate the application of object-oriented development techniques ( as an appropriate technique ) to develop a database through prototyping. Further, recommendations are made for considerations by the Bureau to fully benefit from the solution suggested by way of identifying the prototype database into fully operational database.Item The Application of OCR Techniques to the Amharic Script(Addis Ababa University, 1997-05) Alemu, Worku; Biru, Tesfaye (PhD)Nowadays, it is becoming increasingly important to have information available for examination and manipulation in digital format, and Optical Character Recognition (OCR) is being recognized as one of valuable instruments in this respect. OCR systems take optical images of a handwritten or printed material, and by recognizing the characters that make up the material, automatically convert the text in the material into digital format for further processing and manipulation - thereby bypassing the labour-intensive and error prone as well as time consuming process of keying. While the use and application of OCR systems seems to have been well developed in languages that use scripts based on Latin, Chinese, Japanese, Bangia, to mention but a few, there is not as yet any effort in this direction for the Amharic language. This study is an attempt to approach the development of an Amharic OCR system by drawing experience elsewhere - to investigate the extent to which suggested OCR algorithms to work with other scripts would apply to recognizing Amharic characters. To this end, algorithms suggested for use in other languages are reviewed from published literature. The Amharic writing system is described in terms of size, shape, style, etc. Algorithms of general appeal to the Amharic character recognition are selected. Experimentation with a step-by-step segmentation and recognition based on topological features of Amharic characters is presented. Recommendations are also made to further the experiment and enhance the performance and applicability of the selected algorithms.Item Applying Interface Agent Technology to Selective Dissemination of Information (Sdi) User Profile Management: The Case of ilralerts(Addis Ababa University, 1998-05) Yimam, Dawir; Biru, Tesfaye (PhD)SDI services and systems are most important components of information services. However, they are also widely reported to be resource intensive and time-consuming for both the service providers and users. As a result, despite the immense contribution they are proved to provide, they are rarely implemented in developing countries whose information systems and services are mainly characterized by paucity of resources. In particular, previous studies have repeatedly stated that the creation and maintenance of the user profile is the most problematic component. This thesis research investigated in to the problems of this particular component with the view to apply an advanced Artificial Intelligence techno logy known as interface agents which are a type of software systems generally called agents. To this end, the SDI user profile management problems are analyzed taking a case SDI system known as Aerialists of the International Livestock Research Institute (ILRI), Addis Ababa. This has enabled to discover numerous ways in which agent technology could bring about improvements on SDI user profile management. Furthermore, the review of the literature has shown that the technology is already being employed in systems which are similar to, or enhancements of, SDI systems like those commonly known as filtering systems that are developed for Internet resources. Drawing on these findings, an agent-oriented modelling of the user profile management process has been made and some of the basic methods and procedures suggested in this model have been demonstrated by developing a prototype agent. Using this agent, some small scale testing of the performance of the suggested methods have been made taking few test data from ILRIAlerts users. The prototype development and tests have indicated the possibility of employing agent methods to SDI user profile managementItem Automatic Morphological Analyzer for Amharic: An Experiment Employing Unsupervised Learning and Auto segmental Analysis Approaches(Addis Ababa University, 2002-06) Bayu, Tesfaye; Biru, Tesfaye (PhD); Leyew, Zelealem (PhD)Item Automatic sentence Parsing for Amharic Text an experiment using probabilistic Context free grammar(Addis Ababa University, 2002-07) Alemu, Atelach; Biru, Tesfaye (PhD)Natural Language processing, as a field of scientific inquiry, plays an important role in increasing computers capability to understand natural languages, the language by which most human knowledge is recorded. Works in the area of Natural Language Processing try to design and implement computer programs that can understand natural language and act appropriately on the information contained in the text or utterance. Enabling computers to understand natural language involves extraction of meaning from natural language sentences. And one of the steps in this process is sentence parsing. Sentence parsing, which is also called syntactic parsing, is the process of identifying how words can be put together to form correct sentences and determining what structural role each word plays in the sentence and what phrases are subparts of what other phrases. A sentence parser outputs a parse structure that could be used as a component in many applications including semantic analysis, machine translation, information storage and retrieval of textual data etc. Today, parsers of different kinds (e.g. probabilistic, rule based) have been developed for languages, which have relatively wider use nationally and/or internationally (e .g. English, German, Chinese, etc). The same story is not true for Amharic, the working language of the Federal Government of Ethiopia, and one of the major languages of Ethiopia (Bender et ai, 1976) since to the best of my knowledge, there are no sentence parsers of any sort that process this language.Sentence parsing, which is also called syntactic parsing, is the process of identifying how words can be put together to form correct sentences and determining what structural role each word plays in the sentence and what phrases are subparts of what other phrases. A sentence parser outputs a parse structure that could be used as a component in many applications including semantic analysis, machine translation, information storage and retrieval of textual data etc. Today, parsers of different kinds (e.g. probabilistic, rule based) have been developed for languages, which have relatively wider use nationally and/or internationally (e .g. English, German, Chinese, etc). The same story is not true for Amharic, the working language of the Federal Government of Ethiopia, and one of the major languages of Ethiopia (Bender et ai, 1976) since to the best of my knowledge, there are no sentence parsers of any sort that process this language. This study, thus, attempted to develop a simple automatic parser for Amharic texts/sentences to address the need for developing systems that automatically process the Amharic language. In the study, the Inside Outside algorithm with a bottom up chart parsing strategy has been used. The probabilistic context free grammar has been used as a grammatical formalism to represent the phrase structure rules of the language. A small sample corpus was selected from sentences in the language, and has been used to serve as a training and test set. The sample was then hand parsed, automatically tagged, and was used as a corpus to extract the grammar rules and assign probabilities. The thesis, in short, describes processes of automatic sentence parsing using a combination of probabilistic and rule-based reasoning. It describes the whole process from manually parsing simple sentences to developing a prototype and conducting an experiment with it. The results obtained using the small manually parsed corpus seems to encourage further research to be launched, especially with the aim of developing a full-fledged Amharic sentence parser.Item A Computer-Assisted Financial Information Support System For National Urban Planning Institute (NUPI).(Addis Ababa University, 1996-06) Mama, Alemtsehay; Biru, Tesfaye (PhD)Ability to produce information that has improved qualities such as relevance, timeliness and accuracy is among the main objectives of financial information systems including NUPI'S finance system . But the system is facing difficulties in producing s uch outputs. Most of users requirements are not being met by the syst em . As a result this study is conducted. The study documents what and how the existing system carries out its activities. Current system problems are identified. Obj ecti ves that should be met by a proposed system are set . Various ways of meeting those obj ecti ves are investigated. Finally a so lution is proposed. Attempt is also made to demonstrate how the solution over comes the major problems of the existing syst em . The outcome of this s tudy works not only fir NUPI but also for organisation s and institutions that are financed by MOF and using single entry Government accounting system.Item A Computer-Assisted Financial Information Support System For National Urban Planning Instiute (Nupi).(Addis Ababa University, 1996-06) Mamo, Alemrsehay; Biru, Tesfaye (PhD)Ability to produce information that has improved qualities such as relevance , timeliness and accuracy among the main objectives of financial information systems including NUPIS finance system. But the system is facing difficulties in producing such outputs. Most of users requirements are not being met by the system . As a result this study is conducted. The study documents what and how the existing system carries out its activities. Current system problems are identified. Objectives that should be met by a proposed system are set. Various ways of meeting those objectives are investigated. Finally a solution is proposed. Attempt is also made to demonstrate how the solution over comes the major problems of the existing system. The outcome of this study works not only fir NUPI but also for organizations and institutions that are financed by MOF and using single entry Government accounting system.Item Data Mining Application in Supporting Fraud Detection on Mobile Communication: The Case of Ethio-Mobile(Addis Ababa University, 2005-01) Gebreselasie, Jember; Biru, Tesfaye (PhD)The application of data mining methods and tools that can help to explore large quantities of data generated by the Call Detail Record (CDR) of telecommunication switch machine will be the art of the day to address the serious problems of telecommunication operators, The CDR consists of a vast volume of data set about each call made and it is a major resource of data for research works to find out hidden patterns of calls made by customers in addition to the typical use for bill processing activities, More importantly, data mining techno logy has enabled telecommunication companies to utilize this generic source of data for different kinds of customer relation ship management and marketing activities including fraud detection and prevention Strategies The methodology used for this research had three basis steps these were collection of data, data preparation and model building and testing, The required data set was selected and extracted from the billing data set of Ethio- mobile, Neural network data mining technology were employed to build and test the model s, The c1ata mining mcthod used in this research work have proved to yield comparably suJ'fi cient results for practical u s~ as far as supportive mechanisms are employed fo r the misclass ification of fraudulent customers as non-fraudulent and vice versa, Due to this fa ct the telecornmunication operators should take serious and intensive follow ups for unusually high frequency and long duration of international calls made by some possibly fraudulent customers customers However these lection of representative sample data for the whle database of the problem under consideration needs a series consideration and care to proply sdcct the training data set. Otherwise due to the rare occurrences of possibly fraudulent calk the training of representation data set will be a tedious activity In this research work, the researcher has proved that the CDR is a major resource of or crucial knowledge about customers of Telecomm citation Corporation, Beside that the number of calls made and the duration of each call should be traced to proper Know unhealthy customers of the corporation, Finally it is a rich field for research for other interested researchers to make further and in-depth research to this area ,Item Design and Development of an in-Hose Database System For Archival and Manuscript Records Management: The Case of Institute of Ethiopian Studies.(Addis Ababa University, 1998-06) Semeon, Getahune; Fisseha, Sisay (PhD); Biru, Tesfaye (PhD)Ethiopian Study is an area study. "Area study" is a field of study that falls within the scope of a group of studies which are multi-disciplinary in nature. The Manuscript and Documentation Center is one of the major units under the Library at the Institute of Ethiopian Studies that provide important historical information to researchers of area study, to students, staff, foreign scholars, etc. The materials reflect the cultural heritage of the Ethiopian people. The collection constitutes at the Center unpublished materials of various types, parchments, photographs, slides, postcards, microfilms/fiches etc. The Center has information handling and provision problems due to mainly the unique type of collection and the existing manual system of preparing finding aids to this unique collection. It is believed that a computer based information system can help a lot in solving the problems of Manuscript and Documentation Center. In particular, by enhancing the accessibility of the manuscript and archival collection to users through providing indexes, query searching, selective searching by field or key word, and allowing multiple access. To this end the existing system was analyzed based on extensive interviews conducted with the staff, document analysis and personal observation. User groups were identified and their requirements were surveyed using questionnaire. Based on the findings, a system was designed using an object oriented approach. Finally, a prototype was developed using BASE for Windows in order to implement the proposed database solution, test it and draw a recommendations for improvement. Among the problems identified in the existing system are the lack of properly organized finding aids, poor organization or management of the collection, poor housing, no systematic recording of in-house collection management (i.e. accession, exhibition and restoration), lack of trained manpower and problem of security.Item Designing Networked Information System For Construction Companies Using an Object-Oriented Approach: A Case Study of Sur Construction S. Co.(Addis Ababa University, 2000-05) Tafere, Tesfaye; Biru, Tesfaye (PhD)Most system development methods, at least in principle, have two impol1ant components, namely a modeling language and a process. The modeling languages are graphical notations for expressing designs and provides easy communication and documentation. The process of the method shows how to do the development process This thesis uses the Rumbough's OMT (Object Modeling Technique) for the process of the analysis and designing of information SUPPOI-t system for SUR Construction S Co. Equipped with this object oriented methods, the thesis focuses on identifying the requirement of the company, department by dependent. And then a special focus is given to the material resource of the company. It has been found that almost every department is concerned with the material resources, and as such it is a decisive object that determines the success or failure of the company. Therefore, the necessary information and operations concerned with this important material has to be identified and designed so that every department is provided with the kind of information the department requires for decisions making. Many system development effol1s are made with very little or no regard to the very impol1ant aspect of development process, namely, the documentation. Although system development efforts are made in SUR very recently it appears no sufficient documentation for the requirement analysis of the company has been made. As a result, each is an islands of desperate attempts, that failed to come up with a system that can provide an accurate, timely, and relevant in formation that the management of SUR needs for decision making and control of operation. This thesis uses the UML as ways of communicating and documenting the requirement analysis of the company. Risk and uncertainty are fact of life. SUR are faced with a multitude of different risk and uncertainty issues. But the company approaches these unpredictable circumstances of business based on the whims and experiences that can be subject to variation and errors. Hence it is good to collect the experience of the domain experts of the company on the area and formalize and organize the decisions by clearly identifying the important variables both controllable and uncontrollable and describe them through a model. And use this model to forecast how each risk circumstance behaves, to make a better decision that otherwise would have been costly. To this end, this paper explores the potential of models in bidding for a new project with competitors.Item Developing a User Friendly Courseware: A Case Study with Inventory Management(Addis Ababa University, 1995-05) Molla, Alemayehu; Tadesse, Taye (PhD); Biru, Tesfaye (PhD)These days, the use/implementation of computer in education is becoming increasingly attractive. There is now a general understanding among workers in this area that the application of computers in education can be categorized as TUTOR (CAI), TOOL and TUTEE . The TUTOR application (which is the subject of concern here) in particular, relates to the use of computers as educational delivery systems where as the TOOL and TUTEE applications refers to the use of computer for performing certain tasks and for programming respectively. A typ ical eAl/TUTOR system consists of a COURSEWARE (a specially designed software for classroom instructional use), along with the hardware and system software required to run such software. As with any other software, courseware development requires analysis of the requirements first, in particular, answer to such basic questions as When to use a course ware? how to design (develop) a user friendly course ware and what factors to consider in the design of such a course ware. This study attempts to address these questions by way of demonstrating the design and development of a user friendly course ware using the available resources at SISA. A literature survey of definitions, types and design considerations of watercourse along with the main reasons behind the use and current development trends is presented. Among the development models suggested for the use in developing watercourse, the ISD model isItem Flight Revenue Information Support System for Ethiopian Airlines(Addis Ababa University, 2000-05) Mikael, Gobena; Biru, Tesfaye (PhD)Ethiopian Airlines is a profit-oriented business organization whose objective is to provide the maximum value to its customers, consistent with the need to make some return on each transaction. One of the major primary activity at the airline is Sales. In addition, because conU11erci al organizations only survive by identifying and satisfying the market, Marketing Services is also regarded as a major primary activity. This study focuses on the revenue process within the Sales and Marketing operation. In particular, it aims at understanding the critical business functions and processes involved in the flight revenue process, to identify and assess the avai labi lity of revenue data elements and develop a model for Ethiopian Airlines that wi ll support information on revenue realized by flight and forecast revenue by flight; accurately and timely. Ethiopian Airlines has numerous state-of-the-art application systems and as a result retains a vast amount of data in its different databases. However, it has fai led to make good use of this data and has not been ab le to lise it to create competitive advantage. As a result, the revenue information model has been developed using data mining techniques. In particular, the neural network model was used to train, test, validate and develop the prototype model. The ultim ate objective being to find out the suitability of data mining applications to the Ethiopian Airl ines problem.Since the scope of the study is limited to a single organization, the major method that has been used to assess revenue information needs of users is case study; implemented through interv iews (planned discussion), questionnaires, observation and document analysis. After reviewing the various areas that are affected by the Sales and Marketing operation; Sales, Scheduling, Pricing, Revenue Management and Airport Operations have been identified as the critical functions in the revenue process. As a result the foc us of the study has been on these functions. Survey results reveal that of the 5 most important infonn ation required by the concerned airline managers, revenue related information ranks on top with 31 % of respondents ranking it first. In addition, 84% of respondents rate fl ight revenue information as either one of the most or the most critical infOim ation, 88% as either very or extremely strategic, and 94% as one that would provide opportunity to gain competitive advantage. The major revenue related data elements identified during the study are advanced booking data, post departure data, schedule data, and revenue data. These revenue related data elements are available within the existing system, but are scattered in the various application systems. Over one year's historic advance booking data is available, over two years' post-departure data is avail able, and historical fl ight revenue data since April, 1997 is available.After selecting a suitable software to build a revenue information model, the revenue related data elements identified were collected for 8 flights and a comprehensive testing was conducted. The test included 6 different experiments using the back propagation network and radial basis fll11ction neural network models, 3 different sets of independent variab les and a multitude of trai ning parameters. The experiments produced 327 different models which were compared and evaluated and finally one was selected to represent the revenue information model. The developed model, with an average of33-37% error rate, is only a preliminary or initial step towards, hopefully, more detailed work in this area. I am confident that through a selection of more fields and with more hi storical data, the error may be able to be reduced to users' requirement of 5-10%. It is, therefore, my belief that this research has some contribution to further research in this area. It has been ab le to successfully demonstrate that data mining applications can be an alternative approach to build information systems; especially for complex problems having vast amount of data and high interaction among non linear variables. Others can pursue similar research using different types of data mining applications, including other neural network models. I hope that some of the problems I encountered and the methodologies I used will help to shed light and guide others undertaking similar studies.Item Flight Revenue Information Support System for Ethiopian Airlines(Addis Ababa University, 2000-05) Mikael, Gobena; Biru, Tesfaye (PhD)Ethiopian Airlines is a profit-oriented business organization whose objective is to provide the maximum value to its customers, consistent with the need to make some return on each transaction. One of the major primary activity at the airline is Sales. In addition, because commercial organizations only survive by identifying and satisfying the market, Marketing Services is also regarded as a major primary activity This study focuses on the revenue process within the Sales and Marketing operation. In particular, it aims at understanding the critical business functions and processes involved in the flight revenue process, to identify and assess the availability of revenue data elements and develop a model for Ethiopian Airlines that will support information on revenue realized by flight and forecast revenue by flight; accurately and timely. Ethiopian Airlines has numerous state-of-the-art application systems and as a result retains a vast amount of data in its different databases. However, it has failed to make good use of this data and has not been able to use it to create competitive advantage. As a result, the revenue information model has been developed USll1g data mining techniques. In pm1icular, the neural network model was used to train, test, validate and develop the prototype model. The ultimate objective being to find out the suitability of data mining applications to the Ethiopian Airlines problem. Since the scope of the study is limited to a single organization, the major method that has been used to assess revenue information needs of users is case study; implemented through interviews (planned discussion), questionnaires, observation and document analysis. After reviewing the various areas that are affected by the Sales and Marketing operation; Sales, Scheduling, Pricing, Revenue Management and Airport Operations have been identified as the critical functions in the revenue process. As a result the focus of the study has been on these functions. Survey results reveal that of the 5 most important inforn1ation required by the concerned airline managers, revenue related information ranks on top with 31 % of respondents ranking it first. In addition, 84% of respondents rate flight revenue information as either one of the most or the most critical information, 88% as either very or extremely strategic, and 94% as one that would provide opportunity to gain competitive advantage. The major revenue related data elements identified during the study are advanced booking data, post departure data, schedule data, and revenue data. These revenue related data elements are available within the existing system, but are scattered in the various application systems. Over one year's historic advance booking data is available, over two years' post-departure data is available, and historical flight revenue data since April, 1997 is available. After selecting a suitable software to build a revenue infol111ation model, the revenue related data elements identified were collected for 8 fliFghts and a comprehensive testing was conducted. The test included 6 different experiments using the back propagation network and radial basis function neural network models, 3 different sets of independent variables and a multitude of training parameters. The experiments produced 327 different models which were compared and evaluated and finally one was selected to represent the revenue infol111ation model. The developed model, with an average of 33-37% error rate, is only a preliminary or initial step towards, hopefully, more detailed work in this area. I am confident that through a selection of more fields and with more historical data, the error may be able to be reduced to users' requirement of 5-10%. It is, therefore, my belief that this research has some contribution to further research in this area. It has been able to successfully demonstrate that data mining applications can be an alternative approach to build infol111ation systems; especially for complex problems having vast amount of data and high interaction among non linear variables. Others can pursue similar research using different types of data mining applications, including other neural network models. I hope that some of the problems I encountered and the methodologies I used will help to shed light and guide others undertaking similar studies.Item A Generalized Approach to Optical Character Recognition (OCR) Amharic Texts(Addis Ababa University, 2000-05) Meshesha, Million; Biru, Tesfaye (PhD)These days research in Optical Character Recognition is popular for its application potential in banks, post offices, insurance, and other governmental and non-governmental organizations. Other application areas include library automation and natural language processing. As Amharic is the working language of Ethiopia and used as a means of communication by most governmental and non-governmental organizations, there is a huge collection of document and processing that could benefit from OCR system. To this end, since recent times, research in the area of Amharic OCR system has been undertaken at SISA. The present research is a continuation with the aim of improving the performance of the system under investigation at SISA in recognizing characters written in different font types. To this end, feature-based approach was considered after thoroughly studying features of Amharic characters. Algorithms for thinning and feature extractions were reviewed from literature. An attempt was made to implement some of these algorithms so as to see their performance on Amharic text printed in different typeface s. Previous algorithms implemented for segmentation (stage-by-stage segmentation) and feature extraction/detection (tree-based topological features extraction teclmique) are incorporated with some modification to complete the Amharic OCR. The system is then tested on sample Amharic documents of actual cases (written in Agafari, Washra and Visual Geez) and test results obtained for each of the case is repOt1ed. Recommendations are also drawn to highlight areas of further research so as to improve the current work and incorporate other features to Amharic OCR system.Item Internet as an Information Service: Opportunities in Ethiopia and Connectivity Conditions a Case Study of Potential Internet Users in Addis Ababa(Addis Ababa University, 1996-05) Adane, Fanta; Chowdhury, G.G (PhD); Biru, Tesfaye (PhD)Connectivity enhances activities being done by individuals, stimulates interaction and enables access to knowledge and resources. In general, connectivity facilitates quick economic, Social and political transformation. In view of this fact, this study was conducted to assess the opportunities and connectivity Conditions of the Internet in Ethiopia. Questionnaire survey and interview were used to conduct the survey and SAS was used to analyze the collected data. Stratified sampling was followed in assessing the potential Internet users in different sectors of the community . One hundred twenty one (121) sample of e-mail users of the PADISNet were selected from Private / Individual Academic and Research, NGOs, governmental, and Interaction al sectors and 90 of them have replied the The questionnaire The findings of the survey indicated that a large number (81 . 1%) of the respondents are aware of the Internee, its benefits and pitfalls as well . It was found out from the survey that majority (67 . 8% ) of the respondents require e-mail service, a large number (88 . 9%) of them need textual information and 78 . 9% responded that they are interested on research data on the Internet . In addition, the highest (84 . 5%) number of potential Internet users believed that it is the right time for the country be connected to the Internet and 53 . 3% of them responded that they are capable of covering the cost of the Internet. Based on the above conclusion, the following recommendations are made to sustain a successful Internet service in EthiopiaItem The Marketing of Science and Technology Information Products and Services in Sub-Saharan Africa: A Case Study of the Ethiopian Science and Technology Commission (Estc)(Addis Ababa University, 1997-05) Ngulube, Patrick; Biru, Tesfaye (PhD)This thesis addresses the topic of the marketing of information services and products in Sub Saharan Africa with special reference to Ethiopia. It examines the broadened concept of marketing and how the concept has been applied in the library and information services(LIS) world in general and Sub Saharan Africa in particular. The marketing concept which is a philosophy that advocates a user oriented organization and views users as the lifeblood of the business is also explored. Various marketing models are considered and a network analysis/scheduling model is suggested. The steps in model building were market research and portfolio analysis. The research revealed that marketing strategies are not applied in most LIS in Sub Saharan Africa. Marketing planning is conspicuously absent. Much of what passed for marketing in LIS in Sub Saharan Africa was a variety of discrete activities rather than a planned and coherent programmed Although information services and products exist their utilization is very low. Utilization of information seems to depend on the awareness of its existence. For users to be aware and utilize the information resources , and thus maximize the return on investment marketing planning is suggested. The illustration of the network analysis/scheduling model for marketing planning at the Ethiopian Science and Technology Commission is given.Item The Marketing of Science and Technology Information Products and Services in Sub-Saharan Africa: A Case Study of the Ethiopian Science and Technology Commission (ESTC)(Addis Ababa University, 1997-05) Ngulube, Patrick; Biru, Tesfaye (PhD)This thesis addresses the topic of the marketing of information services and products in Sub Saharan Africa with special reference to Ethiopia. It examines the broadened concept of marketing and how the concept has been applied in the library and information services(LIS) world in general and Sub Saharan Africa in particular. The marketing concept which is a philosophy that advocates a user oriented organisation and views users as the lifeblood of the business is also explored. Various marketing models are considered and a network analysis/scheduling model is suggested. The steps in model building were market research and portfolio analysis. The research revealed that marketing strategies are not applied in most LIS in Sub Saharan Africa. Marketing planning is conspicuously absent. Much of what passed for marketing in LIS in Sub Saharan Africa was a variety of discrete activities rather than a planned and coherent programme. Although information services and products exist their utilisation is very low. Utilisation of information seems to depend on the awareness of its existence. For users to be aware and utilize the information resources , and thus maximize the return on investment marketing planning is suggested. The illustration of the network analysis/scheduling model for marketing planning at the Ethiopian Science and Technology Commission is given.