Master Dissertations

Permanent URI for this collection

http://192.168.18.118:4000/handle/20.500.12661/349

Browse

Now showing 1 - 20 of 94

Analysing the effect of visibility and scintillation on free space optical communication: a case of Dodoma and Dar es salaam
(The University of Dodoma, 2018) Chiyaba, Teck Kinte
Free space optical communication (FSO) is receiving a lot of attention as an alternative last mile wireless communication system. This is because of its high data rate, large bandwidth and easy installation than the traditional radio frequency (RF) and fiber optic communication systems. However the medium through which the signal travels is the atmosphere. This medium is affected by the prevailing weather conditions. FSO is prone to hash atmospheric conditions such as wind, temperature, humidity which induce scintillation. Rain, fog, snow, aerosols are the other conditions which induce visibility and affects the communication channel. In order to establish the feasibility of the technology while taking into account these impediments this study calculated visibility and scintillation attenuation for Dar es Salaam and Dodoma based on data obtained from Tanzania Meteorological Agency (TMA) and weather underground (WU). The attenuation coefficients were also calculated and input into the simulation package called OptiSystem 7 to simulate the bit error, Q factor and eye diagrams. The results of the simulation have shown that Dar es Salaam has higher FSO attenuation that Dodoma. It has also been shown that Dodoma weather conditions provide better FSO communication than Dar es Salaam. The transmission using the study parameters shows that one can transmit for a distance of eight (8) km for Dar es salaam and above twelve (12) kilometre for Dodoma. A comparison of the two modulation schemes has shown that NRZ is the best modulation scheme that could be used for the two cities and October is best month for transmission in Dodoma. This technology is feasible under the parameters that were used in this study and therefore is recommended for adoption.
Analysis of the impact of soil electrical conductivity on magnetic inductiion-based communication between underground and aboveground network devices
(The University of Dodoma, 2018) Mahenge, Emmanuel
Magnetic induction (MI) is a physical layer technology that is effective for underground wireless communications. The underground communications involve three major communication links, namely underground to underground (UG2UG), underground to aboveground (UG2AG), and aboveground to underground (AG2UG). AG2UG and UG2AG links are special in a sense that they cross two different media, which are soil and free space. While this is the case, the soil electrical conductivity (EC) is proved to be among the factors that affect MI underground communication. Therefore, this study looked at the impact of the changes in soil EC on UG2AG and AG2UG links. Channel models were reviewed and the model for communication between underground and aboveground devices that included the attenuation factor of soil EC was proposed. Experiments were also done in four different fields; at Agricultural Research Institute – Uyole, The University of Dodoma, Tanzania Research and Career Development Institute and Gairo area. These fields were used as they had different soil types and varying soil ECs. Experiments were done using GNU radio, HackRF One transceivers, MI coils and soil EC tester. The results of the experiments were analyzed using MATLAB software. The results showed that the AG2UG and UG2AG had a similar performance in terms of power received. Also, the increase in soil EC value increases the rate of attenuation of magnetic field signal. However, changes in EC were seen to have more impact on higher frequency (433.92 MHz) than on lower frequency (13.56 MHz) due to skin effect. Thus, 13.56 MHz can provide a preferable performance for UG2AG and AG2UG communications. Based on the results of the experiments, a hybrid MI and electromagnetic waves architecture for underground wireless networks was proposed.
An application of mobile phone in pests control and diseases management for vegetable crops: a case of tomato production in Kilolo district
(The University of Dodoma, 2016) Kitundu, Rahim S.
The study has intended to investigate the application of mobile phones in information dissemination regarding pests control and diseases management for vegetable crops with a consideration of tomatoes productions in Kilolo district. The study aimed at bridging the gap that exists between farmers and agricultural officers. The study involved two wards, and was mainly qualitative supported by quantitative research paradigms to collect and analyse data from tomato farmers, ICT expertise and agricultural officer. The study revealed that, pests and diseases were the main causative of low tomato harvest. Lack of proper means of information dissemination between farmers and agricultural officers; caused the damage to be even enormous as farmers searched help from other farmer by visiting each other‟s unaffected farms instead of using interactive mobile phones application as they proposed. Data were collected through questionnaires, interviews and documentary reviews and analyzed qualitatively and quantitatively with the support of Statistical Package of Social Sciences. The collected data, acted as system requirements for designing, developing and testing of ToMatoApp mobile phones application system. ToMatoApp was developed by using waterfall model methodology. System modules were developed and tested by using android studio emulator and smart phones. Mobile phones have potential roles in information dissemination on pests control and diseases management for tomatoes production. However, there is a need of mutual co-operation between the government, agricultural stakeholders, ICT experts and tomato farmers to increase tomato production, quality and livelihood of farmers
Assessing the impact of lack of network documentation in higher learning institutions: Case of University of Dodoma
(The University of Dodoma, 2014) Kachemela, Seif J
Network complexity continues to grow as enterprises add and remove different communication technologies. Software tools that identify network elements, provide logical and physical network views, track network changes, and provide accurate documentation for all network components which are needed and should be employed in the organizations. It has been observed that most of the Higher Learning Institutions (HLIs) in Tanzania facing serious problems in monitoring and managing network because of neglecting documentation. This study intended to assess the impact of lack of network documentation in Higher Learning Institutions in Tanzania. Specific objectives were; to identify factors that contribute to negligence of network documentation in HLIs, to identify the effects of lack of network documentation in HLIs and to develop a prototype that will be adopted in higher learning institutions to document network. Methods used in data collection were in depth interviews and observations. The study population comprised of system administrators from the University of Dodoma. Purposive sampling was employed to get the sample size of 22 system administrators. The researcher discovered that awareness and negligence are among the factors which contributed to the lack of network documentation in HLIs. In addition, researcher found out that troubleshooting and managing network without network documentation is difficult. Finally the researcher develops prototype and use NetBrain Workstation to document UDOM network.
Assessment of the suitable modulation scheme for broadband power line communication system
(The University of Dodoma, 2012) Chuwa, Maria G
In Power line communication the information are sent via electrical power lines. The power lines are characterized by multipath fading and frequency selectivity due to impedance mismatch of the channel and impulsive noise. Therefore there should be a way to increase performance of the system. In this dissertation OFDM and MCSS modulation techniques were used to enhance PLC system performance. Bit Error Rate (BER) performance of OFDM and MCSS was analyzed and system efficiency was enhanced by the application of coding techniques. The performance of these modulation techniques were also analyzed with different loads’ impedances. The performance of the system was assessed by the commonly used Bit Error Rate against Signal to Noise Ratio (SNR) diagrams and the comparison of OFDM and MCSS systems was assessed by using the frequency response obtained from the PLC multipath channel model for different number of paths. For both coded and uncoded systems it was observed that at low load impedances the performance of MCSS was much better at higher signal to noise ratio than that of OFDM by a gain of 4dB. But at lower SNR OFDM had a better BER performance compared to MCSS. Additionally in the case of high load impedances the performance of OFDM was better than that of MCSS for systems with a gain of 12dB.
Automation of electric power supply systems: TANESCO case study
(The University of Dodoma, 2012) Joseph, Richard
An electric power system includes a generating subsystem, a transmission subsystem and a distribution subsystem. An electrical power network is larger and more complex so that, automation of power system is required to monitor and control an electrical power network. Automation of power system mean use of computer to collect data along power network, transferring data to a distribution control centre, displaying the data and carrying out analysis for control decision and improvement in system operation. Currently, TANESCO has feedback concerning the real-time (automatic system) operation at the level of the generation subsystem to high voltage level, but has no feedback concerning the real-time status of their networks at medium and low voltage level. Therefore, the use of automation system in TANESCO entire network will facilitate reduction of number of activities along electrical power network, number of staffs, operational costs and provide better service to consumers. This study was aimed to determine the concepts of automation of management of power systems from generation points to end users in which researcher included the challenges involved in automation of power systems. Also, researcher described communication systems identified for automation of power systems and monitoring systems of automation of power systems. Challenges involved in automation of power system were established in this study. Also, four communication systems (PLCs, Optical Fibers, Satellite Communications and Wireless Communications) for automation of power systems were identified but, two communication systems (PLCs and Optical fiber) were selected as most effective communication systems for automation of power systems. Monitoring tools such as ETAP software and PSS/E software were identified; PSS/E tool was selected because of its availability. With the aim of determining the concepts of automation of management of power systems, researcher conducted simulation using PSS/E tool and the parameters were used during simulation are power flow and short circuit condition. This study employed case study design; Studying different documents and Interview guide are the methods were used to collect data related to communication systems, monitoring tools and challenge involved in automation of power systems. It is recommended for utility company to use automation of power systems. The use of power system automation reduces numbers of activities, staffs and failure can be observed timely to effect action timely and this means that, makes quick fault detection, isolation, service restoration and therefore better services to consumers.
Barriers on effective utilization of web2.0 technologies for knowledge sharing in Tanzania higher learning institutions
(The University of Dodoma, 2015) Ndaba, Alex Zakaria
Instructors of Higher Learning Institutions (HLI) are under increasing pressure to use ICTs to impart students with the knowledge, skills and attitude they need to compete academically in the twenty first century. Due to this fact, HLIs need to migrate from instructor-centered lecture based instructions to student-centered interactive learning environment. Despite the open nature of Web2.0 for e-learning initiatives and the increasing population of students interacting with the Web for accessing educational information, preliminary investigation and literature reviews have shown that instructors of HLIs in Tanzania are lagging behind on using these platforms for sharing knowledge with the students. This study attempt to investigate the barriers facing the adoption and utilization of Web2.0 technologies for knowledge sharing in Tanzania Higher Learning Institutions. The population of the study was the students and instructors of The University of Dodoma, St. Johns University of Tanzania and College of Business Education. Observation and questionnaires were used as research tools to collect data from 201 respondents who were proportionally selected. Quantitative data was analyzed using SPSS. Views of instructors and students towards utilization of Web2.0 for knowledge sharing were collected to constitute the qualitative data. The findings of this study showed that uncontrolled knowledge sharing, security, intellectual property rights lack of skills and privacy issues pertaining the use of Web2.0 for knowledge sharing are among the barriers towards effective utilization of Web2.0 technologies in Higher Learning Institutions. Also it was found that, while most of the instructors do not use Web2.0 for sharing their knowledge with students, most of the students of Higher Learning Institutions rely on search engines like Google in search of materials for their studies. Finally basing on vi study findings, the researcher develops a prototype that can serve as a platform for enhancing knowledge sharing in Tanzania Higher Learning Institutions.
Cloud computing framework for resource sharing among Tanzanian higher learning institutions
(The University of Dodoma, 2020) Omary, Bally
Higher Learning Institutions world-wide have been depending highly on Information Technology for their business requirements and service delivery. They need substantial investment for procuring and maintaining software as well as hardware. Due to the growing needs and financial crisis, universities are facing challenges in acquiring high-performance computing facilities to support not only advanced analysis but also scientific researches. This study intended to develop a cloud computing framework which can be used for sharing high performance computing resources of higher learning institutions. The study deployed both qualitative and quantitative approaches. Literature review and observation were used as data collection methods. The study furthermore used Microsoft Excel to analyse collected data. After the designing process, the proposed cloud framework was evaluated using two performance metrics. The parameters used for assessment were: response time and processing time. The study findings revealed that, increasing data centres’ power and speed is not always a factor for improving cloud computing performance but sometimes can add more cost. Furthermore, the findings discovered that the use of closest data centres provides better performance than placing them in separate locations. The study suggested the developed cloud computing framework model to be deployed and used by Tanzanian higher learning institutions to reduce procurement and running costs of high-performance computing resources.
Comparative analysis on reception of digital video broadcast-terrestrial: A case study of Arusha city
(The University of Dodoma, 2014) Iddi, Ibrahim
Tanzania has been among the earliest countries in Africa to employ Digital Video Broadcast-Terrestrial (DVB-T) System. The deadline for Analogue Switch Off (ASO) was set to be in 2015, but Tanzania decided to have phased switch off since 2010. DVB-T is a new system in Tanzania and hence the extent of quality of the signal needs to be established. This research compares the reception of DVB-T second generation spatial signal strength as offered by Star Media Limited, Agape Associates Limited and Basic Transmission Limited multiplexing companies in five locations of the Arusha City. The research involved two phases, first was field measurement of DVB-T signals using four different antennas and a spectrum analyzer, and secondly, customers perception on quality of the signal they receive was established using a guided interview. Measurements have shown that overall DVB-T signal strength was quiet strong in Arusha city centre, which allows spectrum analyzer to have a set point of - 10 dBm. In other areas, overall DVB-T signal strength was weak which forced to have the spectrum analyzer set point lowered to -30 dBm. The research also found all of the Digitek and Agape decoder customers were satisfied with the quality of DVBT signal they were receiving. 9 out of 10 customers of Continental decoders and 4 out of 5 Tanzania Broadcast Corporation (TBC) decoders were also satisfied with quality of signal as offered by their MUXs. Research results recommend to Multiplexers to either increase power of their existing towers in allowable ranges or to increase number of towers to strengthen signal reception in remote areas and areas in low altitude. Gap filler can be employed in town to those areas obstructed by high rise structures.
Comparative study of pagerank and hits algorithms for reciprocal link prediction in online social networks
(The University of Dodoma, 2020) Pallangyo, Brian Somi
Online Social Networks (OSN) provides active space for digital human interaction and are used daily. Human engagement is reflected by exploiting the dynamics of OSN, where the fundamental problem is to infer future interactions on the network, called link prediction. Most studies have employed classical algorithms which consider node similarity but neglected the link analysis algorithms which consider topological structure. This study focused on the comparative study of predicting reciprocal interaction from para-social interaction using algorithms. Particularly, this study selected PageRank and HITS, which are considered famous link analysis algorithms with high order heuristics. Network simulation was performed to understand the performance of the algorithms when used to predict reciprocal link formation by employing machine learning techniques. For the experiment, two datasets were used to ensure the reliability of the results. Initially, the publicly available secondary dataset of Twitter was used followed by primary dataset crawled from Mayocoo, both of which are directed networks. The resulting networks from both datasets adhere to power-law distribution. Resource allocation was used as the baseline for the study after outperforming Adamic-Adar, Jaccard Coefficient, and Preferential Attachment. The result of this study showed that both PageRank and HITS surpassed the baseline in performance of prediction. Thus, PageRank has an accuracy improvement of 1.8% with precision and recall of 4.8% and 1.1%, respectively. Furthermore, this improvement comes with a balance of 3% (f1-measure). When HITS is used, there is an improvement accuracy by 5%, with 15.1% (precision), 7.9% (recall) and 11.5% (f1-measure). These empirical results demonstrate that HITS outperforms PageRank in prediction performance. Also, the results from the computational test showed that PageRank uses less computational resources compared to HITS. This study suggests the use of link analysis algorithms over classical algorithms for reciprocal link prediction in OSN. Furthermore, the use of HITS is recommended when prediction performance is vital compared to computational cost, otherwise, PageRank in cases were computational resources are minimal.
A comparative study on performance of support vector machine and convolution neural network on Tanzania sign language translation using image recognition
(The University of Dodoma, 2020) Myagila, Kasian
Sign language has been used by Speech impaired people for communication purposes. Despite being an effective form of communication for speech impaired people, still there is a challenge for people who are unaware of sign language especially those with no such impairment to communicate with speech impaired people. Since Sign Language is a visual based language, several machine learning techniques have been used in sign language translation for better performance results. However, sign languages are different and no study has been found in Tanzania Sign Language, which is a language used by speech impaired people in Tanzania. Moreover, no study has revealed whether there is significant difference in performance between Support Vector Machine and Convolution Neural Network despite the fact that literature show that both have significant performance in different sign languages. This study aimed at comparing the performance of Support Vector Machine and Convolution Neural Network on translating Tanzania Sign Language through image recognition. The study employed Tanzania Sign Language images as datasets whereby 30 words were chosen from the context of education. The study used dataset of 3000 images that were taken using a camera. To reduce the dimension of datasets, the study adopted Principal Component Analysis to perform feature extraction. Furthermore, the study employed a Combined 5x2cv F test to compare the techniques to determine the significant difference in the performance of the algorithms. The findings revealed that both techniques have significant rate of both accuracy, precision and recall. Convolution Neural Network scored 96% in all of the parameters while the SVM with Histogram Oriented Gradient feature scored similar rate in precision but lag on recall and accuracy by 1%. Additionally, the results of using Combined 5x2cv F test yield a p-value of 0.0258 which shows that there is a significant difference in the performance of the two techniques when used to translate the Tanzania Sign Language. Therefore, this study recommends the use of Convolution Neural Network since it has high accuracy and it can provide a significant higher rate of performance compared to Support Vector Machine.
Credibility judgement and verification behavior on online information
(The University of Dodoma, 2016) Kilombele, John
Judging the credibility of information that is found online is very crucial, because on the internet there are platforms which contain misinformation, though they may look trustworthy. Editorial policies regarding the majority of information posted online are non-existent, unknown or ambiguous. Thus, developing the skills to evaluate web-based information is crucial for internet users. There is a need to understand how information seekers in Tanzania do prepared for this responsibility and to understand weather they have trouble determining the credibility of online information. Thus, this study aims at finding out to what extent do users care about judging the credibility of information that is found online. Methods of data collection used in this study were a structured questionnaire and Literature review. Questionnaire was used to explore the extent to which user care about judging credibility of information found online, whereas literature was used to describe the verification strategies used. Findings indicate that information seekers in general do not care judging credibility of online information meaning that users do not often conduct critical evaluation on information that is found online. Therefore, this study recommends that learners and people who use internet as their major source of information must be trained and be prepared to use their skills and effort in judging the credibility of information that are found online, because on online platforms there is a possibility of encountering low quality information.
Data integrity risk assessment of local government revenue collection information system in Tanzania
(The University of Dodoma, 2018) Msigwa, Bilali Bahati
This study grounded at assessing the data integrity risks of local government revenue collection information system (LGRCIS). It was observed through the CAG report that the LGRCIS had weaknesses which were causing data mismatch. Thus, the main purpose of this study was to assess the risks of LGRCIS typically concentrating on data integrity challenges that cause those data mismatch between the LGRCIS and POS machine. To address the problem the researcher went through covering three specific objectives, which were to identify the causes of data mismatch in LGRCIS, determine the frequency of occurrence and their resulting impacts and proposing risk reduction mechanisms which will help to reduce those impacts and improve system efficiency. A sample size of 75 respondents from five district councils of Iringa Region was used. Data collection was carried out by the use of questionnaires as the main tool and face to face interview. Qualitative and quantitative data were gathered and analyzed by using Statistical Package for the Social Sciences (SPSS) and Microsoft Excel Office for plotting tables, pie and bar charts. The study identified both human and technological factors which were the main challenging threats in the system that caused data mismatch as well as the frequency of occurrence and their resulting impacts that hindered the revenue collection process. The study concluded by proposing the risk reduction mechanisms that could help the system work efficiently. In this sense the researcher proposed that, the central government to set a contract with telecommunication companies to provide reliable and high speed Internet connection for all points located at remote areas with network coverage problem. Also, ICT officers should use pre-paid system for recharging Internet bundle for POS machine to avoid delay and training to system users should be conducted frequently together with system performance should be frequently audited by external auditors.
Design and implementation of virtual library system: a case of Mbeya University of Science and Technology
(The University of Dodoma, 2017) Daud, Shadrack
In this study, the problem of library service provisional using traditional library systems was examined. In Tanzania and some parts of the world in general, it has been revealed that many institutions still they are using physical library materials and moreover, users of the library has to attend a physical point for the sake of accessing library materials. For library users, this has led to minimal utilization of library services due to number of obstacles caused with this system of providing library services and to mention few they include collision on sharing materials such as few numbers of books, delay on accessing the services due to few numbers of staff, and improper record management of library materials. After going through a survey on documentation and using a questionnaire to study the current level of library services of a traditional library service provision, using a specific case study, the researcher has tried to suggest the use of virtual library tool to examine the performance improvement on library services provision. The tool was constructed using software development methodologies and was installed locally and using web servers within the institution and users were added to the system and were allowed to login using authorization credentials and were permitted to access electronic library materials, accessing a collection of online libraries and a number of hyperlinks of library materials. Statistics were made on the users' frequency of material accessibility, links visited, books and journal ordered or borrowed. The results have shown there is a significant increase of the users visited the virtual library system as compared to the normal traditional library system. It was also revealed that many users were able to access electronic library material. This has signified that a virtual library tool is an important tool for improving library service accessibility in organizations.
Design of an in-home multicarrier broadband power line communication system for Tanzania homes
(The University of Dodoma, 2012) Victor, Crallet
Indoor broadband power-line communication (BPLC) networks with different number of branches, different terminal load impedance, different line length and impulsive noise using OFDM were considered. Earlier investigations have shown that the power-line channel suffers from multipath effects along with impulsive noise. In this dissertation the OFDM system was designed based on Anatory et al model using the impulse responses of typical Tanzania indoor power-line networks. It was observed that the maximum delay spread for such networks is 2μs, making the number of sub-channels for OFDM to be 512 with a cyclic prefix of 64 at a bandwidth of 16.5MHz. Bit error rate (BER) performance of the OFDM system under multipath effects was analysed based on the simulations of the four Tanzania indoor power-line networks at different load impedances with number of branches and branch length dictated by a particular home power-line network. It was observed that the performance of the OFDM system is affected by number of branches and load impedances. To maintain the bit error performance an average SNR per bit gain of 28dB is needed so that sustained communication is still available in all the four Tanzania indoor power-line networks. The OFDM system performance at the fixed load impedance within the same house was different indicating that the number of branches has the influence on the performance.
The design of wireless sensor network for smart honey bee farming monitoring
(The University of Dodoma, 2016) Joseph, John J
Honey bee farming has been one of the earliest cultural and economic activity practiced by human being, dating back 15000 years ago in north of Africa in the times of the pharaohs. Honey is one of the most important natural products in the world which is produced by bees. However for decades honey bee farming monitoring in developing countries like Tanzania have depended much on cultural and traditional method of monitoring which cost time, money, and resources. Furthermore, efficient tools for monitoring can facilitate in promoting sustainable method that counter this problem at a greater extent. Emerging of technologies like wireless sensor network (WSN) can be used to provide inexpensive management, coordination and intelligent management network for monitoring the bees and helping the farmers in improving the production of honey. This dissertation investigates the application of WSN in smart honey bee farming, identify the proper requirements for wireless sensor actuator network for smart honey bee farming and lastly to analyzes on the performance of the system. Different tools and appropriate open source WSN technologies are identified for the proposed system network, with the aim of encouraging optimal performance and allowing low power consumption to the system. Furthermore the society is expected to benefit from this study by helping the farmers in promoting and enabling effective monitoring method that will be applied in improving honey production in Tanzania.
Designing a model for an integrated hospital Information systems in Tanzania: a case of Dodoma and Manyara regions
(The University of Dodoma, 2017) Arusha, Kinto
This study has intended to design a model for an Integrated Hospital Information System in Tanzania, and was conducted at four hospitals from Dodoma and Manyara regions respectively. A sample of 90 respondents participated in the study from a purposive and non-purposive sampling. A Statistical Package for Social Science (SPSS) was used to analyze the quantitative data. Findings have revealed that Integrated Hospital Information System (iHIS) in Tanzania requires infrastructure, system development, data sharing and interoperability and unique identification as well as information security and privacy. Strong security measures must be observed before integration, during integration and after integration of HIS system. The study not only should design an architectural model for iHIS but also the design has considered the fact that, most patients’ records which were captured by the HIS deployed in various hospitals in Tanzania did not have either common unique identities or biometric data, which could be used to identify a patient’s record throughout the country. An integrated hospital information system which could enable sharing of patients’ information between the hospitals is an important tool for healthcare providers to provide continuous treatment to patients, understand the background of the disease, understand how the patient was diagnosed and knows the treatment the patient is on. This study also, recommends the following to be the principle for achieving the iHIS. The systems should be as flexibility as possible in order to allow further growth and its expansion; interface engine must be constituted with technologies that suit for integration such as XML, JSON and REST as well as a patient having a unique identification number which can be used in all integrated systems
Designing and implementation of an inferential engine prototype for degree-program recommendation
(The University of Dodoma, 2018) Mahenge, Isaac
In this study students’ admission forms and the Internet were used to create datasets in order to design, implement and test a prototype for an inferential engine for degree programme recommendation. The said engine is a Machine Leaning (ML) tool to be used by students for selection of degree programmes. The dataset had 17 features which represented ordinary and advanced level performances, category of schools of admission and student gender. Data in the dataset was unevenly distributed in nine classes, whereby 80% was used for training with 10-fold cross validation and 20% was used for testing of the seven selected ML algorithms. The ML algorithms which were selected for this study were Decision Tree (ID3), Nearest Neighbor, Support Vector Machine (RBF kernel), and Bagging Classifier. Others were Random Forest, Adaptive Boost, and Neural Network (MLP). Random Forest outperformed the other ML algorithms with an accuracy of 66%, with Mean Absolute Error of 11.93. RF attained precision, recall and F-measure of 66% each; Cohen’s kappa and MCC of 60% each; Log Loss of 29%; and Hamming Loss of 34%. The study recommends the educational governing institutions to use a well-formed evaluation and record keeping system to enable easy tracking of student performance.
Detection of DDoS attacks and flash events occuring simultaneously in network traffic using deep learning techniques
(The University of Dodoma, 2020) Mihanjo, Carl Eginald
Recently, the advancement of technology and internet contributes to the increase of the network traffic over the globe. It improves digital services delivery over the global network such as online shopping, television, and streaming. However, as digital services become one of the de facto applications over the internet, a number of attacks on them have been increasing which raise security concerns. Some of the major attacks are Distributed Denial of Service (DDoS) and Flash Events (FE). One hand DDoS attacks mainly focus on disrupting the legitimate users to access the internet. On the other side, FE occurs when there is a rapid growth of legitimate users that access the service over the internet and overload the system. DDoS attacks and FE have similar behaviour however, they need different countermeasures. The major challenge lies in detection the attacks especially when DDoS and FE happen simultaneously. The study proposed a model to detect the FE and DDoS attacks when occurring simultaneously in network traffic using deep learning techniques with three different hidden layers and two optimizers. Validations of the models were tested with data from the real network traffics and the model with high performance was selected which was a model with three hidden layer and Adam optimizer. The result shows a proposed model achieved a good accuracy of99% and less than 1% false alarm.
Determination of factors affecting security and operation performance into management of land information system: a case of Dar es salaam
(The University of Dodoma, 2016) Kiyungi, Haruna I.
The aim of this study was to determine factors that affect the security and operation performance into the Management of Land Information System (MOLIS) at the Ministry of Land, Housing and Human Settlement Development (MLHHSD). The researcher used the case study design; whereby both qualitative and quantitative approaches were used for gathering data for this study. The study investigated on the potential security issues and operation details of the network, hosts/servers, and applications such as MOLIS. Questionnaires were used for gathering important details from ICT experts, system users as well as customers. The observation method was used for collecting potential details from system users as well as system operation behaviours. Moreover, the researcher employed the use of security vulnerability scanners such as; GFI LanGuard 2012 and Nmap 7.01 for the collection of potential security technical details of network, system and computers. The Statistical Package for Social Sciences (SPSS) version 20 and Microsoft excel were used during analysis for the data, whereby, both qualitative and quantitative analysis methods were used. The use of descriptive statistics method was employed in order to determine the different percentages of the respondents. Then, qualitative data was analyzed with the use of content analysis to get meaningful information. Findings from the study reveal several factors that were contributing in security and operation performance degradation in MOLIS; whereby factors such as weak system administration, system design flaws, poor system configurations, system backdoor activities, bureaucracy, corruption, were some of the mentioned factors associated to. Moreover, the study used Adobe vulnerability severity rating for categorization and assessment for the potential vulnerability revealed during the study.

Browse

Browsing Master Dissertations by Title

Results Per Page

Sort Options