Computer Science

Permanent URI for this collectionhttps://etd.hu.edu.et/handle/123456789/76

Browse

Now showing 1 - 20 of 39

DECISION FRAMEWORK FOR THE USAGE OF CLOUD TECHNOLOGY IN ETHIOPIA HIGHER EDUCATION INSTITUTIONS
(Hawassa Unversity, 2019-10-06) SELAM DESALGNE
The rapid technology advancements are always creating new opportunities and a new way of working. Cloud Technology is being popularizing across the world especially in academic institutions. It is not a new technology but rather a new delivery model for information and services using existing technologies. The paradigm has been recognized recently as key enabling efficient and effective technological services that will reshape the delivery and support of the educational services. This study is conducted on public Ethiopian Higher Educational Institutions to explore the critical determinants that influence the adoption of the Cloud Technology. Despite the fact that cloud computing offers great deal of opportunities, its adoption exacerbated with lack of standards and relative lack of general framework created dilemma for the institutions how to approach the cloud adoption. An exploratory study is carried out. This research work proposes TOETAD conceptual framework according to the Technology Organization Environment (TOE) model, Diffusion of Innovation (DOI) theory and Technology Acceptance Model (TAM) with added Decision Maker Context to the model. Adoption determinants for the technology will be examined through the lens of integrated model. The framework factors were identified by critically reviewing studies found in the literature together with factors from the industrial standards within the context of Ethiopia Higher Education Institutions. Data is collected by online questionnaire survey with IT managers, lectures, E-learning coordinators and Team Leaders from selected 17 Ethiopia Higher Educational Institutions with a total 103 respondent. On the other hand the proposed frame work is evaluated by an expert to validate the framework. The result also helps to encourage the Public Higher Educational Institutions in Ethiopia to understand the nature of the problem, increase their awareness about factors to be considered while adopting the cloud computing
ETHIOPIAN COFFEE BEAN DETECTION AND CLASSIFICATION USING DEEP LEARNING
(Hawassa University, 2020-06-02) GETABALEW AMTATE
Ethiopia is the homeland of Coffee Arabica. Coffee is the major commodity export which covers the highest income source of foreign currency. In addition to this, Coffee has a great role in social interaction between peoples and the source of income for the coffee-producing farmers. Ethiopian coffee beans are distinct from each other in terms of quality based on their geographical origins. Classification and grading of those coffee beans are based on growing origin, altitude, bean shape and color, preparation method and others. However, the quality of the coffee beans is determined by visual inspection, which is subjective, laborious, and prone to error and this requires the development of an alternative method which is precise, non destructive and objective. Thus, the objective of this research is to design and develop a model that characterizes and identifies coffee beans of six different origins of Ethiopia (Jimma, Limmu, Nekemte, Yirgacheffe, Bebeka, and Sidama). Coffee beans for this research are collected from the Ethiopian Coffee Quality Inspection and Auction Center (ECQIAC). Image processing and the state-of-the-art deep-learning techniques were employed to automatically classify coffee bean images into nine different class: washed Limmu, unwashed Limmu, washed Sidamo, unwashed Sidamo, washed Yirgacheffe, unwashed Yirgacheffe, unwashed Jimma, unwashed Nekemte, and washed Bebeka. A total of 9836 coffee bean images were used to train, validate and test the CNN model. We have compared the classification result of the model trained on different dataset sizes and hyperparameters. The model was trained on 80% of the dataset, validated on 10%, and tested on 10% of the colorful coffee bean images, with batch normalization has scored 99.89% overall classification accuracy and 0.92% generalization log loss. In conclusion, the result of the study shows that CNN is an effective deep learning technique in the classification of Ethiopian coffee beans
COMPUTATIONAL MODEL FOR WOLAYITTA LANGUAGE SPELLING CHECKER
(Hawassa Inversity, 2020-08-10) RAHEL SHUME
Spelling checker systems are built, and researches are conducted worldwide to meet the needs of different languages. In Ethiopia, spelling checker researches were carried out for only Amharic language. These works have paved the way for researches on other languages like Wolayitta. A computational spelling checker is proposed and adopted for the Wolayitta language in this research. Word level spelling checker were built based on an edit distance algorithm and language model based open source tools and platforms. A dictionary database was constructed from 13,313 words from Wolayitta lexicon dictionary and 20,000 words from Wolayitta Bible. The database was then clustered into training and testing. The statistical model accept input from the user then check whether it is available in dictionary or not if it is available it will do nothing if not it gives suggestion. The testing was done in two phases where the first is error detection rate and the second evaluation is error correction. Tests were carried out on the input data from the user, and accuracy of 94.57% was achieved for spelling error detection, while accuracy of around 90.98 % was retrieved for the spelling suggestion model. The results were promising, and further researches can be entertained as per the recommendations made by the researcher
DEVELOPING IMAGE-BASED ENSET PLANT DISEASE IDENTIFICATION USING CONVOLUTIONAL NEURAL NETWORK
(Hawassa University, 2020-11-07) UMER NURI MOHAMMED
Nowadays, decline in food plant productivity is a major problem causing food insecurity to which plant disease is one of the factors. Early identification and accurate diagnosis of the health status of food plants is hence critical to limit the spread of plant diseases and it should be in a technological manner rather than by the labor force. Traditional observation methods by farmers or domain experts is perhaps time-consuming, expensive and sometimes inaccurate. Based on the literature, the literature suggests that deep learning approaches are the most accurate models for the detection of plant disease. Convolutional Neural network (CNN) is one of the popular approaches that allows computational models that are composed of multiple processing layers to learn representations of image data with multiple levels of abstraction. These models have dramatically improved the state-of-the-art in visual object recognition and image classification that makes it a good way for enset plant disease classification problems. For this purpose, we used an appropriate CNN based model for identifying and classifying the three most critical diseases of enset plants: - enset bacterial wilt, enset Leaf spot, and Root mealybug diseases. Enset is one of a major source of food in the South, Central and Southwestern parts of Ethiopia. A total of 14,992 images are used for conducting experiments including augmented images with four different categories; three diseased and a healthy class obtained from the different agricultural sectors stationed at Hawassa and Worabe Ethiopia, these images are provided as input to the proposed model. Under the 10-fold cross-validation strategy, the experimental results show that the proposed model can effectively detect and classify four classes of enset plant diseases with the best classification accuracy of 99.53%, which is higher than compared to other classical deep learning models such as MobileNet and Inception v3 deep learning models
MASTERS OF SCIENCE IN COMMUNICATION ENGINEERING AND NETWORKING
(Hawassa Unversity, 2021) SHIMELIS ABATE BELAY
Currently, number of wireless technologies deployed to connect daily activities of human being in different ways and systems. Among those technology Wireless Local Area Network (WLAN) the one which plays crucial role. Hence, it is delivered by use of 2.4GHz and 5GHz frequencies. Due to the number of users increase the two main problem became challenges which are spectrum scarcity and throughput. To solve the challenge there are several types of research are done and are going to be done related to 60GHz Millimeter Wave Frequency for WALN. In this thesis, coverage and capacity performance comparison of 60 GHz channel capacity over Rican fading channels with 2.4 GHz and 5 GHz for WLAN service to give a better selection for WLAN users in the future. By using with bit error rate (BER) and SNR for small-scale (fast) fading with higher M-ary QAM modulation scheme. As a result, the Rician Channel fading for 60GHz with comparisons of 2.4 and 5 GHz WLAN frequency has high throughputs, which is 60GHz channel capacity is 13.5 times of 5GHz channels, and 54 times channels of 2.4GHz. Therefore, it is more advantageous for high throughput user demands than 2.4 and 5GHz frequencies used for IEEE 802.11’s Standards. 60 GHz distance coverage, relatively time less half coverage of than 2.4GHz frequency and less than by around 7-meter time less than 5GHz frequency coverage. Hence, the shorter the coverage of 60GHz give an advantage to best candidate for frequency re-use to solve spectrum scarcity
QUERY EXPANSION FOR AFAAN OROMO INFORMATION RETRIEVAL USING AUTOMATIC THESAURUS
(Hawassa University, 2021-03-05) SAMUEL MESFIN BAYU
Recently, the amount of textual information written in Afaan Oromo language is increasing dynamically. Likewise, the need to access the information also increases. But, it is difficult to retrieve and satisfy one`s own information need, because of the inability of the users to formulate a good query and the terminological variation or term mismatching among the world of readers and the world of authors. Hence, query expansion is an effective mechanism to reduce term mismatching problems and also to improve the retrieval performance of IR systems. The idea behind query expansion is to reformulate the user’s original query by adding related terms. In this study, an automatic Afaan Oromo thesaurus is constructed from manually collected documents. After the text preprocessing tasks are performed on the document corpus, the preprocessed words are vectorized in multidimensional space by using Word2Vec`s skip-gram model. In which, words that share similar context have similar vector representation. Then cosine similarity measure was applied to construct the thesaurus. A one-to-many association approach was employed to select expansion terms. Hence top five terms that have the highest similarity score with the entire query were selected from the thesaurus and added to the original query of the user for query expansion. Then the reformulated query was used to retrieve more relevant documents. Experimentations were performed to observe the quality of the constructed thesaurus and the effect of integrating query expansion into the Afaan Oromo IR system. The result shows that the constructed thesaurus generates related terms with average relatedness accuracy of 62.1%. On the other hand, the integration of query expansion registered performance improvement by 14.3 % recall, 2.9 % F-measure, and performance decrement of 5.5% for precision
DESIGN OF A MORPHOLOGICAL GENERATOR AND ANALYZER FOR SIDAAMU AFOO VERBS USING FINITE-STATE TRANSDUCER
(Hawassa University, 2021-03-07) KITAW AYELE GESUMA
Sidaamu Afoo is an official language of the newly formed sidaama national state. It is one of the under-resourced languages of Ethiopia. There are very small attempts of study in the morphological perspective of the language. The morphological study is a low-level activity that paves the way for the development of high-level NLP applications. This study was interested in the design and implementation of a morphological generator and analyzer for the language verbs. The finite state transducer is used to design the verb morphological generator and analyzer system. Lexicon formalism is used for designing method and foma programming language as implementation toolkit. Forty-one distinct linguistic rules were used to form those variate word forms. The experimental result shows that out of 74,934 total generated words, 94.8% of words are correctly generated and analyzed by the system. Only 5.2% of words generated are invalid and wrongly analyzed. The factors of the problem are optional replacement rules and compiler ignorance of the glottal consonant of the language. Strict demarcation boundary for optional replacement rules and alternative techniques for glottal sound management left for future work.
A URL-based Phishing Attack Detection and Data Protection Model
(Hawassa University, 2021-09-10) Yonathan Bukure Racho
Internet users are increasing rapidly in an uninterrupted way that is influencing the way of living. Every day billions of websites are accessed over the globe to facilitate different usage to people. This positive reinforcement is also resulting in internet abusing by hackers for their benefits. Most of the time internet abusing has experimented with over mobile phones or emails. The users are victimized by those abuse even without knowing that they are misused by hackers. Social engineering has become the tool for the hacker to manipulate users psychologically to reveal secret information. Phishing is a kind of social engineering attack with the potential to do harm to the individual or overall organization. Cybercriminal called Phisher comes up constantly in contact with individuals with creative ways to compromise the secret assets. Phishers uses the malicious URLs that are embedded over the webpage with severe threat and appears legitimate. When user clicks these links, redirects to malicious webpage where attackers ask some secrete information by misguiding user. Such kinds of attacks must be properly addressed. This thesis is focused on URL based phishing detection and data protection against such kind of attacks. Thus, the contribution of this thesis is divided into two phases that are: (1) URL based phishing attack detection, and (2) Protection of individual/organization assets. For the first phase, this thesis explored and implemented four machine learning algorithms like Decision tree, Random Forest, Naive Bayes, and Logistic Regression. Further performances of these algorithms are evaluated and compared against training and testing dataset. Based on performance result obtained, the best algorithm is recommended. For the second phase, thesis proposed a data protection model using a hybrid encryption method that combined AES and RSA algorithms. This model ensures the confidentiality of information assets as well as protect them against various kind of attacks. Overall proposed work is implemented in the Python programming language. The phishing detection phase concluded that Random forest outperforms and gave the highest accuracy of detection after important feature selection as compared to other algorithms. Results analysis conclude 96.89% and 99.06% detection accuracy over testing and training dataset respectively in Random forest. Similarly, the data protection phase encrypted and decrypted the data files very fast i.e., within few milliseconds and ensured the confidentiality of data in transit
EXPLORING A BETTER FEATURE EXTRACTION METHOD FOR AMHARIC HATE SPEECH DETECTION
(Hawassa University, 2021-10-08) YESUF MOHAMED YIMAM
Hate speech is a speech that causes people to be attacked, discriminated, and hated because of their personal and collective identities. When hate speech grows, it will cause death and displacement of peoples from their homes and properties. Social media has the ability of widely spreading hate speech. To solve this problem, various researchers have studied many ways to detect social media hate speeches that are spreading in international and local languages. Because the problem is so serious, it needs to be carefully studied and better addressed in a variety of solutions. The previous studies detect a speech as hate speech, based on the frequency (occurrence) of a word in a given dataset; this means it does not consider the role of each word in a given sentence. The main purpose of this study is to design a method that can generate hate speech features from a given text by identifying the role of a word in a given sentence, so that hate speech can easily be distinguished from other forms of speech in a better way. To do this, various researches related to this study have been studied and reviewed. This study created a new feature extraction method for Amharic hate speech detection. The created model needs a training and testing dataset, so that posts and comments, which are posted on 25 popular Facebook pages, have been collected to build the dataset. Whether a speech is hateful or not, should be determined by the law that prohibits hate speech. So that, using different filtration methods, datasets that contain religious, ethnic, and hate words are collected and given to law experts, to annotate it manually. The law experts labeled 2590 datasets into three classes; Religion-hate, Ethnic-hate, and Non-hate. After dataset preparation, a new feature extraction method, which can distinguish hate speech from other speech, is developed. The new feature extraction method and other feature extraction methods that are used in other related studies are implemented and computed with three machine learning classification algorithms: SVM, NB, and RF. The result in different evaluation metrics shows that the new feature extraction method performed better in all combinations of classification algorithms. By using 80% of 2590 labeled datasets as a training set and the rest as a test set, 96.2% average accuracy is achieved using the combination of SVM with the new feature extraction method.
Developing Koorete Part of Speech (POS) Tagger: an Empirical Evaluation of Neural Word Embedding and N-Gram Based Statistical Approaches
(Hawassa University, 2021-12-12) Agegnehu Ashenafi
The Koorete language is spoken by the Koore people in Amaro Kele Special Woreda and in four Kebeles of Burji Special Woreda, Southern regional state. Koorete is written with Latin alphabets (or called ‗Diizo Beyta’ in Koorete language). This means, Latin alphabet is adopted to the language by adding additional combinations of letters for peculiar sounds totaling to 31- consonants (‗Artaxita’ in Koorete), 5-vowels (‗Arxaxita’ in Koorete), and one more symbol. The syntax of Koorete sentence structure is “Subject (Zeere utaade) + Object (efaxe) + Verb (Hanta beyiisaxe)”. This study develops Koorete POS Tagger using the empirical evaluation of Neural Word Embedding and N-gram based statistical POS tagging approaches. Parts-of-speech (POS) tagging is the process of assigning part-of-speech labels/tags to each word from Koorete POS tagset. Neural word embedding are distributed representations of words into vectors applying Bi-LSTM RNN model. N-gram based statistical approach uses probability frequencies of sequence labeling of words from the KPT corpus. Words having similar meanings can be represented similarly, which enable deep learning methods. The behavior of having similar representation orients to the reduction of out-of-vocabulary impact. This means, binary vector |V| dimension reduction. In simple language, word embedding is a language modeling technique which maps words to vectors using Word2Vec package, and would be computed in RNN. This Word2Vec package converts words to arrays of real numbers, and concatenate the original corpus word categories to the generated vectors. Word2Vec has a capability of capturing context of a word (semantic and syntactic similarity) in a document in relation with other words. For the purpose of sequence labeling method and distributed representation, this study uses Bi LSTM RNN by achieving the state-of-the-art POS tagging accuracy and N-gram based statistics approaches in contrast to the more classic approaches. Bi-LSTM handles or adds letter case functions to keep the original letter case information of word. This study applies skip-gram algorithm to encode words into a limited vector space. Because skip-gram model is efficient method for learning high quality vector representations of words from large amounts of unstructured text data. So experiments were practiced on Bi-LSTM RNN model, and N-gram tagger statistical approach. For this, KPT corpus is used about size of 1718 sentences (33220 words), and then divided this corpus into 90% training data and 10% testing data. The experiment on Bi-LSTM RNN word embedding POS tagging approach did better than the N-gram statistical POS tagging approach with the accuracy of 98.53%. Hence, this study solves the problems of (1) no rich resource in NLP applications, (2) Koorete language not having its own KPT corpus and tagsets for NLP applications, (3) the state-of-the art tagging performance algorithms accuracy with other relative languages POS tagging model.
DATA CENTER VIRTUALIZATION FRAMEWORK IN ETHIOPIAN HIGHER
(Hawassa University, 2022-03-04) ABDALLAH HUSEN
Ethiopian higher learning institutions(HLI) data center runs physical data center in which a one application for one server architecture is exercised. Such architecture leads to under utilization and wastage of resources. Using virtualization technology enables lowering the size of the infrastructure, thus resulting in a huge savings on energy and other resources including management time. Cost reduction , energy efficiency and the reduction of a company’s carbon footprint are some of the significant benefits of using virtualization but virtualization technology is a long way from Plug-and-Play to unlock those benefits effectively, an information technology (IT) expert requires the appropriate set of model to be followed. In most cases, organizations have enough resources to move to virtualization without requiring additional budget. Certain researches have been conducted on virtualization especially client virtualization, application virtualization and network virtualization. But, little attention was given for server virtualization. The objective of this research is to explore the current traditional infrastructure practice and to propose data center virtualization framework using design science research methodology (DSRM). The goal is to look for option(s) that brings a preferred solution for University data centers that increases service availability and utilization of hardware. In this work a virtualization framework is proposed, and evaluated using the same services that are currently used in the university. Experiments are carried out to check and compare the resource utilization of both the physical machines and Virtual machines(VMs). By using both the experiment and the analysis result, well-matched virtualization framework was developed. The developed frameworks was tested on three services. Before virtualization each of these services were running in separate physical machine. In our experiment, by consolidating these services using virtualization, the study showed that it is possible to provide the same service using only a single server. The resource utilization of that single server increased as follows: Central processing system (CPU) usage increased from 4% to 17%, physical memory usage increased from 15% of 16GB to 62% of 16GB, and capacity of hard disk space in use was 67.4% up from 32%. The Universities data center will use the proposed well-matched framework to enable them increase their levels of utilization of the servers
CONTEXT-BASED SPELL CHECKER FOR SIDAAMU AFOO
(Hawassa University, 2022-03-04) MEKONEN MOKE WAGARA
A spell checker is one of the applications of natural language processing that is used to detect and correct spelling errors in written text. Spelling errors that occur in the written text can be non-word errors or real-word errors. A non-word error is a misspelled word that is not found in the language and has no meaning whereas a real-word error, that is, the word is a valid word in the language but it does not fit contextually in the sentence. We designed and implemented a spell checker for Sidaamu Afoo that can detect and correct both non-word and real-word errors. Sidaamu Afoo is one of the languages spoken in the Sidaama region in the south-central part of Ethiopia. It is an official working language and is used as a medium of instruction in primary schools of the Sidaama national regional state in Ethiopia. To address the issue of spelling errors in the Sidaamu Afoo text, a spell checker is required. In this study, the dictionary look-up approach with a hashing algorithm is used to detect non-word errors, and the character-based encoder-decoder model is used to correct the non-word errors. The LSTM model with attention mechanism and edit distance is used to detect and correct the context based spelling error. To conduct the experiment, 55440 sentences were used, of which 90% were for training (i.e., 49,896) and 10% were for testing (i.e., 5544). According to the experimental results, for an isolated spell checker, dictionary lookup with hashing achieved an accuracy of 93.05%, a recall of correct words of 91.51%, and a precision of incorrect words of 72.37% for detection. The encoder decoder model achieved a recall of 91.76% for corrections. For a context-sensitive spell checker, the LSTM model with attention and edit distance achieved an accuracy of 88.8%, recall of the correct word of 86.18%, and precision of the incorrect word of 62.84% for detection. It achieved a recall of 74.28% for the correction. The results of the experiment show that the model used to detect and correct both non-word and real-word spelling errors in Sidaamu Afoo’s written text performed well. Finally, to improve the performance of the model, we recommend using additional data set and a state-of-the-art transformer model.
TRAINEE PERFORMANCE PREDICTION MODEL FOR HAWASSA POLYTECHNIC COLLEGE USING KDP
(Hawassa University, 2022-03-05) TIZITA G/SILASSIE
Data mining is the key tool for discovery of knowledge from large data set. It is this technology in most of the educational organization of the world currently helping to know the organization data explicitly and pave the way to produce quality citizens. Unlike other sectors the power of data mining is not much exploited in educational sector. Although there are studies regarding academic performance of students using data mining techniques, they are all about university students. we cant find academic performance of trainee research in Technical and vocational Institute. Thus, the purpose of this study is to develop Trainee performance prediction model for Hawassa polytechnic college. A total of 8200 records with 13 attributes were collected from Hawassa Polytechnic College registrar data set of the past 5 years ranging from 2009 to 2013 E.C. An experiment has been conducted using the Knowledge Discovery Process (KDP) Model using WEKA software version 3.8.4. Four data mining algorithms namely J48 Decision Trees, JRip rules induction, Naïve Bayes and PART with seven experiments (J48 Pruned and Un-pruned decision tree algorithm, Naive Bayes classifier, JRIP Pruned and Un-pruned and PART Pruned and Un pruned) were used to develop trainees performance predictive model. All the experiments were carried out with the same dataset and evaluated with 10-fold cross validation, 80% and 66% split test parameters. The study shows PART Un-pruned 10-fold cross validation test has the highest accuracy with 95.4268% and attributes such as trade/occupation, EGSECE, transcript, level, sex, English, and sector can be used at a time of decision making as they have shown strong prediction power which can help to predict trainees performance. Finally the researcher develop a prototype based on the rules generated from the selected algorithm.
IMAGE BASED BARLEY LEAF FUNGAL DISEASE DETECTION USING CONVOLUTIONAL NEURAL NETWORK
(Hawassa University, 2022-03-06) MAMO GUDISA
Barley is one of the most grown crops in the east Arsi, west Arsi, and Bale zones of Oromia Region. It is one of the main sources of food and income in these and other areas of Ethiopia. However, barley crop was affected by fungal disease which reduces the production and main cause for the economic losses in agricultural industries in Ethiopia. For the betterment of human health, fungal diseases in leaf of barley crops must be controlled and effectively monitored. The earlier researchers have used hand-crafted-features for image classification and recognition with machine learning approach. Nowadays, the development in Deep Learning has allowed researchers to drastically improve the accuracy of object detection and classification. In this thesis, the researcher used a deep-learning approach which is Convolutional Neural Network algorithm to detect fungal disease of barley crop using leaf images collected from Arsi zone of Kulumsa research center and other images captured directly from different Barley farms. The researcher dataset contains two categories of barley crop: leaf rust and normal. The dataset contains 10,224 healthy and diseased images. From this, 80% of the images are used for training and the rest for testing the model. During training, data augmentation is used to generate more images to fit the proposed model. Additionally, many researchers agree that using data augmentation can also increase the performance of the model. The designed model is trained and tested using the collected dataset and compared with two pre-trained convolutional neural network models namely Mobile Net and InceptionV3. The model obtained 99.53% accuracy and it can be used as a practical tool for farmers to protect barley crop, against fungal diseases
DEEP LEARNING BASED FABA BEANS LEAF DISEASES DETECTION AND CLASSIFICATION
(Hawassa University, 2022-03-08) MARTHA MEZGEBU HAILU
Faba bean (Vicia Faba L.) is believed to be originated from the Near East and now days spread throughout the world. It’s one of the most domesticly legume in the world next to chickpea and pea. Ethiopia is the second leading producer of Faba beans next to China in the world. It shares 6.96% of world production and 40.5% with of Africa. Faba bean is grown primarily for its edible seeds that are used for human consumption. It also used for keeping human healthy and sustaining the productivity of the farming system through the fixation of nitrogen. However, most of the time it is affected by different diseases that result in reduction of quality and quantity of the Faba bean production. Those diseases are caused by fungus, virus, and bacteria. Usually Faba bean diseases appear on the leaf, flower, pods, seed, and stem a step by step and makes the crop out of usage. Mainly, leaf of Faba bean is more affected by diseases than other parts. It attacks both inside and outside of the leaves. Leaf plays an important role during the growing period of Faba bean. Without leaf there is no flower, without flower there is no pod, without pod there is no seed. Traditionally, farmers and experts detect and identify plant diseases by naked eyes. This method is inaccurate and expensive, because there are numerous diseases. Detection by using image processing techniques has been more accurate and fast. Therefore, we need to develop automatic deep learning based Faba bean leaf diseases detection and classification model. We designed Faba bean leaf disease model architecture using convolutional neural network for Faba bean leaf diseases detection and classification. CNN become accurate and precise method for the detection and classification of plant diseases. The study can be conducted in the plantation area of Faba bean in Oromia region, Arsi zone, D/Xijo Woreda, from the farmer plantation land particular reference to Bucho Silase kebele, Ethiopia, where the dataset has been collected. Leaves of healthy and infected crops are collected and labeled. Processing of image has been performed with pixel wise operations to enhance the image. It is followed with feature extraction the classification of patterns of captured leaves in order to identify Faba bean plant leaf diseases. Four classifier labels are used as ascochyta blight, chocolate spot- botrytis, rust, and Healthy leaf. The features extracted are fit into the neural network with the dataset was spilt into training set, validation set and testing set, 80%, 10%, and 10% respectively, with the batch size 32 and using Adam optimizer. Faba bean leaf diseases detection and classification model achieved the overall accuracy 99.58%
FOR SIDAMA LANGUAGE USING THE HIDDEN MARKOV MODEL WITH VITERBI ALGORITHM
(Hawassa University, 2022-04-07) BELACHEW KEBEDE ESHETU
The Parts of Speech (POS) tagger is an essential low-level tool in many natural language processing (NLP) applications. POS tagging is the process of assigning a corresponding part of a speech tag to a word that describes how it is used in a sentence. There are different approaches to POS tagging. The most common approaches are rule-based, stochastic, and hybrid POS tagging. In this paper, the stochastic approach, particularly the Hidden Markov Model (HMM) approach with the Viterbi algorithm, was applied to develop the part of the speech tagger for Sidaama. The HMM POS tagger tags the words based on the most probable sequence of words. For training and testing the model, 9,660 Sidaama sentences containing 130,847 tokens (words, punctuation, and symbols) were collected, and 4 experts in the language undertook the POS annotation. Thirty-one (31) POS tags were used in the annotation. The source of the corpus is fables, news, reading passages, and some scripts from the Bible. 90% of the corpus is used for training and the remaining 10% is used for testing. The POS tagger was implemented using the Python programming language (python 3.7.0) and the Natural Language Toolkit (NLTK 3.0.0). The performance of the Sidaama POS tagger was tested and validated using a ten-fold cross-validation technique. In the performance analysis experiment, the model achieved an accuracy of 91.25% for HMM model and 98.46% with the Viterbi algorithm
Improving delay tolerant network buffer management approach for rural area’s health professionals’ information exchange syste
(Hawassa University, 2022-08-06) Mulusew Abebe
Delay-tolerant networks (DTNs) are mobile networks in the field of wireless network which are emphasized to provide end-to-end connectivity in the areas where the networks are not reliable and often susceptible to interferences. Despite the rapid advancement of communication technology, there are still rural places that are not connected to the Internet. Health information exchange between rural area and the urban areas still hampered by in adequate telecommunication infrastructures coverage, intermittent connectivity and absence of end-to-end connectivity. The term Delay Tolerant Network (DTN) is invented to bridged communication gaps that have not been connected to the Internet. In current TCP/IP technology communication is possible only when end-to-end path is available. As a result, the usual Internet and TCP/IP network cannot be valid for some hard environments which are characterized by lack of direct path between nodes, lot of power outages and intermittent connectivity. In this work, the researcher investigated the performance of various delay tolerant network routing protocols and selected MaxProp which is convenient for the proposed framework. Most routing algorithm of delay tolerant network assume the nodes buffer space as unlimited but, it is not the case in reality. As flooding-based routing relies on buffer to have a copy of every message at every node, buffer space has substantial impact on delivery probability. The existing buffer management policies compute in biased way, directed by a single parameter in a random manner while other relevant parameters are completely neglected, resulting in an inability to make a reasonable selection. Therefore, the researcher proposed a reasonable buffer management approach on the situations where there is a short contact duration, limited bandwidth and buffer. The proposed buffer management approach improves buffer availability by implementing three buffer management strategies: scheduling, dropping, and clearing buffers entirely for computing purposes, using three parameters: message type, hop count and time to live. The performance of proposed approach is validated through simulation by using opportunistic Network Environment (ONE) simulator. They were analyzed on three metrics, namely delivery probability, average latency and overhead ratio. The simulation results collected in this thesis shows that when the nodes buffer get constrained the proposed method MaxProp Routing based on Message Type Priority (MPRMTP) perform better than the existing buffer management policy by increasing the message delivery quality and decreasing overhead ratio. However, when there is sufficient buffer space, both MaxProp, and MPRMTP shows comparable performance
A MODEL TOWARDS PRICE PREDICTION FOR COMMODITIES USING DEEP LEARNING: CASE OF ETHIOPIAN COMMODITY EXCHANGE
(Hawassa University, 2022-10-03) SOLEN GOBENA
The development of information technology makes it possible to collect and store large amounts of data every second. Market Enterprises are generating large amounts of data, and it is difficult to use traditional data analysis methods to analyze and predict their future market price. Price predictions are an integral component of trade and policy analysis. The prices of agricultural commodities directly influence the real income of farmers and it also affects the national foreign currency. Haricot bean is produced in many areas of Ethiopia and it is rich in starch, protein, and dietary fiber, and is an excellent source of minerals and vitamins. Haricot bean is also the main agricultural commodity traded on the Ethiopian commodity exchange (ECX) market for the past 10 years. Though there are price prediction works for various crops in Ethiopia and abroad using machine learning and deep learning approaches, price prediction for Haricot bean has not been studied using machine learning as to the best of our knowledge,. The main objective of this study is to develop a price prediction model that can predict future prices of Haricot Bean traded at the ECX market based on time series data. Past 10 years, data has been obtained from the Ethiopian commodity exchange (ECX) with sample dataset size of 12272. Simple linear regression (SLR), multiple linear regression (MLR), and long short term memory (LSTM) were evaluated as predictive models. The results showed that LSTM outperformed other predictive models in all measures of model performance for predicting the Haricot Bean prices by achieving a coefficient of determination (R2 ) of 0.97, mean absolute percentage error (MAPE) of 0.015, and mean absolute error (MAE) of 0.032.
MORPHOLOGY BASED SPELLING CHECKER FOR GEEZ LANGUAGE
(Hawassa University, 2023-03-06) FIKADE CHANE FELEKE
Geez is one of the ancient languages. It belongs to Semitic language family. Many ancient literature and books have written in Geez. Currently, Geez course are offered in various colleges, universities and in some primary schools. However, still developed NLP applications are insufficient for this language. In order to write error free Geez text in less time, spelling checker application is a critical NLP application. Spelling checker is a tool used to detect spelling error in a block of text and gives closer suggestions to the error words. A previous attempt has made to develop a spelling checker for Geez language. This attempt was focus only homophone alphabet interchangeably error. In this study, we proposed morphology based (dictionary lookup and morphological analyzer) approach to Geez language spelling checker. The system have three main compenents.These are text preprocessing, error detection, and error correction. To achieve the objective of this study the researcher builds one main dictionaries that contains Geez language lexicon and morphological feature. The researcher built 6115 unique Geez lexicon and 955 rules had defined. We adopt the Hunspell dictionary and affix file format to design a lexicon (i.e. the knowledge base component) and hashing algorithm for searching. Hunspell is an open source spelling checker tool. It has designed especially for languages that have complex morphology. Finally, the researcher has developed a prototype of a system to test the functionality and performance of the Geez language spelling checker. The accuracy of error detection expressed in terms of precision and recall. In addition, the accuracy of suggestion expressed in terms of suggestion adequacy. Therefore, we got the result of lexical recall 91.9%, error recall 83.7%, lexical precision 97.2%, error precision 62.2% and correct suggestions provided by GLSC 87.5%. The overall performance of the system is 90.05%. We conclude that increase the size of the dictionary and develop well organized rule will increase the overall performance of the Geez language spelling checker.
DETECTION AND CLASSIFICATION OF INDIGENOUS ROCK MINERALS USING DEEP LEARNING TECHNIQUES
(Hawassa University, 2023-03-08) HADIA BERKA WAKO
Ethiopia is undoubtedly a place of riches, with a vast and diverse landmass that is rich in resources. However, less attention has been given in utilizing computing discipline like Artificial Intelligence to solve the current problems in the area of mineral mining in Ethiopia. GUJI Zone is one of Oromia 20 administrative zones blessed with different mineral resources. Despite the fact that mineral has lions share contribution to economy of Ethiopia, little work is done in modernizing the mining industry in Ethiopia especially in empowering small-scale Artisanal community. GUJI is one of the zones following outmoded techniques to identify minerals in mining industry. Rock mineral detection and classification employing conventional methods involves testing physical and chemical properties at both the micro- and macro-scale in the laboratory, which is expensive and time-consuming. Identifying tiny rock minerals and detecting its originality using traditional procedure and techniques takes too much time. Identification of minerals merely through visual observation is often erroneous. To address these problems, a deep learning approach for the classification and detection of Rock Minerals is proposed. The design- science research methodology is followed to achieve the objectives of the research. To conduct this study, 2000 images were collected from Guji’s zone and Mindat.org website. After collecting the images, image pre-processing techniques such as image resizing, image segmentation using roboflow, and image annotation are performed. Moreover, data augmentation is applied to balance the dataset and increase the number of images. This research work focuses on classifying and detecting fifteen types of rock minerals. Based on YOLOv7 deep learning model we have used 70% of the dataset to train the model and 30 % of the dataset to test the performance of the model. Finally, the developed model is evaluated using accuracy, precision, recall, and mAP with other models. Experimental result shows that the accuracy obtained from YOLOv7 is 76%mAP for large objects comparing to other models. Consequently, the pretrained weight of yolov7 achieved a 97.3% accuracy in classifying and detecting with other images