Mekelle University Institutional Repository

Preserving, sharing, and showcasing the scholarly output of Mekelle University since 2000.

Access thousands of theses, dissertations, journal articles, conference papers, and datasets produced by MU researchers and students.
Discover unique collections on Ethiopian agriculture, health sciences, engineering, dryland agriculture, and social sciences.
Comply with open access mandates – all content is freely available with persistent identifiers (Handles/DOIs) and OAI‑PMH harvesting.
Submit your research online – fast workflow with Dublin Core metadata and support for PDF, audio, video, and data files.

Mekelle University is proud to be part of the global DSpace community. Visit the MU Library website for more information or start a new submission.

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 5 of 23

Arts Academy
Book of Abstract
A central archive for Mekelle University’s institutional abstract books from academic and research conferences.
College of Business and Economics
College of Dryland Agriculture and Natural Resources Management
College of Health Sciences

Recent Submissions

Test
(Mekelle University, 2026-02-02) Welegebrial, Tekleweyni Geday
A Machine Learning Framework for Amharic Sentiment Analysis in Social Media Images Using OCR and NLP Techniques
(Mekelle University, 2025-11-28) Halefom Desta Fitsum
In Ethiopia, social media platforms are increasingly used as spaces for public communication, with much of the opinion-rich content embedded in images containing Amharic text. Conventional sentiment analysis methods are designed for plain text and they fail to capture this significant portion of online discourse. Complexity of the Amharic script, scarcity of language processing tools, and limitations in computational resources further restrict automatic analysis of image-based text. So, this study develops an integrated framework that combines Optical Character Recognition (OCR) and Natural Language Processing (NLP) techniques to extract and classify Amharic text from social media images into Positive, Negative, and Neutral categories using machine learning classifiers. A balanced dataset of 600 annotated images was compiled and preprocessed with Open CV for image enhancement and Tesseract OCR for text extraction. The extracted texts underwent different text preprocessing stages, including normalization, character unification, and Stop word removal. Then the preprocessed texts are vectorized using Term Frequency–Inverse Document Frequency (TF-IDF). Four machine learning classifiers Support Vector Machine, Logistic Regression, Naive Bayes, and Random Forest were implemented, and the performance of each classifier were evaluated by different evaluation metrics such as, accuracy, precision, recall, F1-score and confusion matrices. The results from the evaluation metrics showed that SVM achieved the highest accuracy of 86%, Logistic Regression (83%) and Naive Bayes (82%), while Random Forest performed less by achieving 75%. These findings highlights that linear classifiers are suitable for Amharic sentiment analysis under resource-constrained conditions. The study demonstrates the feasibility of integrating OCR and NLP techniques for sentiment analysis of Amharic social media images and provides a solid baseline for future work in morphologically rich language processing.
Development of a Deep Learning-Based Sentiment Analysis System for Tigrigna YouTube Comments
(Mekelle University, 2026-04-11) Mesele Solomon
As social media grows, millions of Tigrigna-speaking users share their thoughts on platforms like YouTube. However, there are no built-in tools for real-time sentiment analysis on these plat forms, and for low-resource languages like Tigrigna, even basic datasets for sentiment analysis are unavailable. This research addresses this problem by developing a real-time sentiment analysis system specifically for Tigrigna YouTube comments using the Design Science Research Methodology (DSRM). A dual-purpose browser extension was developed to facilitate both real-time data collection and live sentiment prediction directly on the YouTube interface. This tool incorporates a Multi-Stage Linguistic Pre-processing Pipeline to distinguish Tigrigna from Amharic, resulting in a gold-standard dataset of 30,353 comments. Because the initial data had very few neutral samples, a human-in-the-loop (HITL) strategy was used, where 1,500 model-predicted neutral samples were manually verified and added. This increased the size of the minority class and improved the system’s ability to recognize neutral comments. The preprocessing pipeline followed a specific order to handle the informal nature of social media text: repeated character reduction, abbreviation expansion, character normalization, and the removal of URLs, punctuation, numbers, and non-Tigrigna text, followed by stop word removal. The final dataset was split into 70% for training, 15% for validation, and 15% for testing to ensure a rigorous evaluation. This was followed by tokenization with a sequence length of 32. This study compared nine different experimental setups using CNN, Bi-LSTM, and Hybrid architectures paired with Word2Vec, Fast Text, and Hybrid embed dings. The results show that the Bi-LSTM model with Fast Text embed dings performed the best, achieving an accuracy of 82% and a Macro F1-score of 78%. The system showed a major improvement in the neutral class while maintaining high performance for positive sentiment. The final system provides users with instant sentiment breakdowns of live YouTube comments, offering a practical tool for real-time monitoring and a significant step forward for Tigrigna natural language processing. This methodology provides a framework that can be used for other low-resource languages. Future work should focus on improving the detection of sarcasm and more complex language patterns in Tigrigna.
Production and Optimization of Briquettes from Sawdust Bagasse blends Using Thermoplastic Waste as a Binder
(Mekelle University, 2026-04-06) Samuel Teklemariam Gaim
he rising need for sustainable, affordable sources of energy has encouraged interest in converting agricultural residues and waste plastics into solid biofuels. This study was conducted to produce and optimize sawdust bagasse blends briquettes. The briquettes were produced using thermoplastic waste, specifically HDPE, as a binder at proportions ranging from 5% to 25%. The bagasse and sawdust were dried, ground, and sieved to a particle size of less than 3 mm, then mixed with softened HDPE. The effects of compaction pressure (5–15 MPa), binder content (5–25%), andsawdust-to-bagasse ratio (25–75%) on briquette density, compressive strength, and key combustion properties, including calorific value, moisture content, ash content, volatile matter, and fixed carbon were evaluated. The optimum parameters obtained from the model were a compaction pressure of 15 MPa, an HDPE binder content of 16.34%, and a sawdust proportion of 51.49%, resulting in a density of 0.9145 g/cm³ and a compressive strength of 2.968 MPa, which provide sufficient mechanical integrity for handling and use. Proximate analysis showed low moisture (5.21%) and ash (3.46%) contents, while ultimate analysis revealed high carbon (53.85%), moderate hydrogen (6.03%), and oxygen (39.97% by difference), with very low nsulfur (0.017%) levels indicating low potential for NOx and SOx emissions during combustion. The optimized briquettes achieved a calorific value of 4107 kcal/kg, reflecting a balance favoring enhanced mechanical durability over the study's maximum observed value of 5061 kcal/kg under other conditions. In conclusion, the optimized briquettes produced from sawdust, bagasse, and HDPE binder show satisfactory mechanical integrity, good fuel characteristics, and low environmental impact, indicating their potential as a sustainable alternative for energy generation while supporting effective management of agricultural and plastic waste.
Optimizing biomethane production from co - digestion organic fraction of municipal solid waste and spent coffee grounds in anaerobic digestion
(Mekelle University, 2026-04-08) Kbrom Teklu
Anaerobic digestion (AD) is a critical method to treat the rapidly growing amount of organic fraction of municipal solid waste (OFMSW) generated due to population growth and the expansion of the urbanization. This study focuses on optimizing biomethane production through anaerobic co-digestion of Organic Fraction of Municipal Solid Waste (OFMSW) and Spent Coffee Grounds (SCG). Laboratory-scale anaerobic digestion experiments were conducted using cow dung as inoculum. This study investigated to identify the optimum mixing ratio, hydraulic retention time (HRT), and inoculum/substrate ratio (I/S) for achieving the maximum biogas production while ensuring a high methane yield, using response surface methodology (RSM) and numerical optimization. The feedstock characterization showed that OFMSW had a total solids content of 20.2% and volatile solids of about 85%, while spent coffee grounds exhibited higher total solids (36%) and volatile solids (88.5%). The carbon-to-nitrogen ratios of OFMSW, SCG, and cow dung were 24.3, 26.4, and 27.7, respectively, indicating favorable nutrient balance for anaerobic digestion. The results showed that co-digestion significantly improved biogas yield and methane content compared to mono substrate digestion. The optimum biogas yields 0.63 L/gVS was achieved at an SCG fraction of 31.5% and 68.5% of OFMSW, HRT of 27 days, and ISR of 0.68. The optimum biomethane yield 0.42 L/gVS was obtained at an SCG fraction of 23.6% and76.4 % of OFMSW, HRT of 31 days, and I/S of 1.4 with methane content ranging from 55–60% and carbon dioxide ranging from 40–45 and less than 1% other trace gases. The findings confirm that co-digestion of OFMSW and SCG is an effective and sustainable approach for renewable energy generation, organic waste reduction, and improved waste management in urban areas.