Browsing by Author "Hailemariam, Sebsibe"

Now showing 1 - 5 of 5
  • No Thumbnail Available
    Item
    Construction of English-Amharic Electronic Subject Dictionary for Science And Technology Terms: An Experiment With Mathema Tical Terms
    (Addis Ababa University, 2001-06) Hailemariam, Sebsibe; Teferi, Dereje (PhD); Gemeda, Demissu (PhD)
    This thesis focuses on the construction of English-Amharic electronic subject dictionary for science and technology mathematical terms Development of the local languages in the third world countries becomes important issue for devel o ping and keeping cultural heritage. The problem and solution to make Amharic a medium of instruction are investigated in the research. The existence of the electronic dictionary for science and technology terms will promote the performance of professionals to translate science and technology documents and to prepare document in the field.An information system is developed that provide basic information and the procedures are described in this document . The performance of the system is also tested using document file and user queries as inputs. The result shows the output information accuracy is inversely proportional to the number of concept bearing words in t he query.
  • No Thumbnail Available
    Item
    Design and Implementation of Automatic Morphological Analyzer for Ge’ez Verbs
    (Addis Ababa University, 2010-11) Berihu, Desta; Hailemariam, Sebsibe
    Morphological analysis is a crucial component of several natural language processing tasks including machine translation, spell-check, speech recognition, dictionary (lexicon) compilation, POS tagging, etc., especially for languages with a highly complex morphology, where stipulating a full lexicon of surface forms is not feasible. Nowadays, analyzers of different kinds have been developed for languages that have relatively wider use internationally. The same cannot be said for Ge‟ez, the classic language of Ethiopia where majority of the Ethiopian indigenous historical, philosophical, ethical, religious, etc., literatures and ancient manuscripts are written with. This study is, thus, an attempt to design a morphological analyzer model for Ge‟ez verbs thereby contributing to the goal of developing a full-fledged NLP application for Ge‟ez. For this purpose, rule-based approaches specifically CV-based and Two-Level Morphology (TLM) are adopted to design the model and to implement the prototype of the analyzer. Besides, the analyzer uses a knowledgebase as a demon while identifying the morphosyntactic features. Finally, algorithms that take into consideration the morphological, morpho-phonological and orthographic properties of Ge‟ez language are developed from scratch and applied, as there are no previous such attempts. The prototype was tested with verbs which are extracted manually by domain experts from all twenty seven New Testament books of the Ethiopic Version Bible. The accuracy of the output generated by the analyzer was compared with the manually prepared analyses of the same verb-set by the language experts at two levels: at features-of-verbs level and at verb level. Accordingly, it is observed that the analyzer has analyzed these verbs with an accuracy of 92.05% at feature level and of 73.98% at verb level. The analysis output comprises the lexeme and all valued morphosyntactic features including affixes together with their syntactical functions, indicated subjects and objects along with their person-gender-number features, tense-mood and stem type of the verb, etc. At large, this research has realized the design and implementation of automatic morphological analyzer for Ge‟ez verbs. Keywords: Ge‟ez verbs, Ge‟ez verbs analyzer, Morphological Analyzer, Ge‟ez Morphology
  • No Thumbnail Available
    Item
    Design and Implementation of Automatic Morphological Analyzer for Ge’ez Verbs
    (Addis Ababa University, 2010-11) Berihu, Desta; Hailemariam, Sebsibe
    Morphological analysis is a crucial component of several natural language processing tasks including machine translation, spell-check, speech recognition, dictionary (lexicon) compilation, POS tagging, etc., especially for languages with a highly complex morphology, where stipulating a full lexicon of surface forms is not feasible. Nowadays, analyzers of different kinds have been developed for languages that have relatively wider use internationally. The same cannot be said for Ge‟ez, the classic language of Ethiopia where majority of the Ethiopian indigenous historical, philosophical, ethical, religious, etc., literatures and ancient manuscripts are written with. This study is, thus, an attempt to design a morphological analyzer model for Ge‟ez verbs thereby contributing to the goal of developing a full-fledged NLP application for Ge‟ez. For this purpose, rule-based approaches specifically CV-based and Two-Level Morphology (TLM) are adopted to design the model and to implement the prototype of the analyzer. Besides, the analyzer uses a knowledgebase as a demon while identifying the morphosyntactic features. Finally, algorithms that take into consideration the morphological, morpho-phonological and orthographic properties of Ge‟ez language are developed from scratch and applied, as there are no previous such attempts. The prototype was tested with verbs which are extracted manually by domain experts from all twenty seven New Testament books of the Ethiopic Version Bible. The accuracy of the output generated by the analyzer was compared with the manually prepared analyses of the same verb-set by the language experts at two levels: at features-of-verbs level and at verb level. Accordingly, it is observed that the analyzer has analyzed these verbs with an accuracy of 92.05% at feature level and of 73.98% at verb level. The analysis output comprises the lexeme and all valued morphosyntactic features including affixes together with their syntactical functions, indicated subjects and objects along with their person-gender-number features, tense-mood and stem type of the verb, etc. At large, this research has realized the design and implementation of automatic morphological analyzer for Ge‟ez verbs. Keywords: Ge‟ez verbs, Ge‟ez verbs analyzer, Morphological Analyzer, Ge‟ez Morphology
  • No Thumbnail Available
    Item
    Ethiopian Finger Spelling Classification: A Study to Automate Ethiopian Sign Language
    (Addis Ababa University, 2008-09) Zerubabel, Legesse; Hailemariam, Sebsibe
    Ethiopian Finger Spelling is one of the communication means used among Ethiopian deaf. It is used to express names and any concepts that do not have sign in Ethiopian Sign Language. To fill the communication gap that exists among the deaf and between the deaf and the hearing, the Ethiopian Finger Spelling is processed using techniques from image processing and pattern processing. In this thesis work, a new attempt is done to design the architecture and select the appropriate techniques for each component of the Ethiopian Finger spelling classification system. The proposed architecture has components for image capturing, feature extraction, hand detection, region of interest segmentation and sign classification. For the tasks of hand detection and sign classification, experiments are conducted to select the appropriate pattern classifier and feature. In addition, the capability of the principal component analysis (PCA) driven and harr-like feature with neural network is tested through experiment. According to the experimental result, neural network pattern processing techniques have high detection and classification rate when compared to the cascaded boosted classifier and template matching techniques for the task of hand detection and sign classification respectively. The overall hand detection rate of 99.43%, 96.59% and 77.27% were obtained using neural network with PCA driven feature, neural network with harr-like feature and boosted cascaded classifier respectively. In addition to this, the overall sign classification rate of 88.08%, 96.22% and 51.44% were obtained using neural network with PCA driven feature, neural network with harr-like feature and template matching respectively. In particular, neural network that use harr-like feature shows better performance for the task of sign classification and neural network with PCA driven feature shows better performance for the task of hand detection. viii Keywords: Ethiopian Sign Language, Ethiopian Finger Spelling, Hand detection and Sign Classification
  • No Thumbnail Available
    Item
    Ethiopian Finger Spelling Classification: A Study to Automate Ethiopian Sign Language
    (Addis Ababa University, 2008-09) Zerubabel, Legesse; Hailemariam, Sebsibe
    Ethiopian Finger Spelling is one of the communication means used among Ethiopian deaf. It is used to express names and any concepts that do not have sign in Ethiopian Sign Language. To fill the communication gap that exists among the deaf and between the deaf and the hearing, the Ethiopian Finger Spelling is processed using techniques from image processing and pattern processing. In this thesis work, a new attempt is done to design the architecture and select the appropriate techniques for each component of the Ethiopian Finger spelling classification system. The proposed architecture has components for image capturing, feature extraction, hand detection, region of interest segmentation and sign classification. For the tasks of hand detection and sign classification, experiments are conducted to select the appropriate pattern classifier and feature. In addition, the capability of the principal component analysis (PCA) driven and harr-like feature with neural network is tested through experiment. According to the experimental result, neural network pattern processing techniques have high detection and classification rate when compared to the cascaded boosted classifier and template matching techniques for the task of hand detection and sign classification respectively. The overall hand detection rate of 99.43%, 96.59% and 77.27% were obtained using neural network with PCA driven feature, neural network with harr-like feature and boosted cascaded classifier respectively. In addition to this, the overall sign classification rate of 88.08%, 96.22% and 51.44% were obtained using neural network with PCA driven feature, neural network with harr-like feature and template matching respectively. In particular, neural network that use harr-like feature shows better performance for the task of sign classification and neural network with PCA driven feature shows better performance for the task of hand detection. viii Keywords: Ethiopian Sign Language, Ethiopian Finger Spelling, Hand detection and Sign Classification