Browse
Recent Submissions
- ItemSelf-adaptive, deep learning model for the detection and classification of Network and Host-level attacks(Strathmore University, 2019-08) Ochieng, NelsonIntelligent computer and network attack detection is the topic of this study. Existing classification and detection models are built using static and old datasets and hence are not self-adaptive to changing network conditions. The models are also mostly evaluated using accuracy alone. Complexity, appropriateness, execution time and understandability are not considered. It is the argument of this study that these would be quite useful and would help in determining the appropriate model that could be implemented in a vendor product. This study collects and curates its own dataset, and therefore investigates various deep learning techniques on it. The outcome is the dataset which can later be standardized as a benchmark, and a comprehensively evaluated self-adaptive model for classification and detection of network attacks.
- ItemMulti-sense agent for proactive screening of alcohol addiction(Strathmore University, 2019-08) Muchiri, WallaceAddiction is a complex condition, a brain disease that is manifested by compulsive substance use despite harmful consequence(s). People with addiction (severe substance use disorder) have an intense focus on using certain substance(s), such as alcohol, drugs or pornography as seen in recent studies, to the point that it takes over their life. Diagnosis of addiction to any substance is usually done in a reactive manner whereby the person is identified once the external symptoms manifest themselves due to the fact that the vast majority of individuals do not seek treatment for their condition. This can be attributed partly to the failure to diagnose early by primary care physicians, the stigma by the society and self-denial by the potential addict since they choose not to seek help until they hit rock-bottom (Zhang & Ho, 2016) thus aggravating the already unwanted situation. Recently there is a shift to the use of ICT techniques such as mobile devices to detect various health symptoms proactively with researchers developing objective mobile datadriven biomarkers for many healthcare conditions such as depression which is highly related to addiction. Furthermore the development of machine learning decision support systems such as the SimSensei Kiosk which is used alongside qualified psychologists in diagnosis of post-traumatic stress disorder have validated the fact that it is possible to successfully implement autonomous diagnosis systems. The aim of this study is to propose an autonomous agent based on the multi-sense framework that can be used on demand for proactive diagnosis of alcohol addiction.
- ItemIntellibot Data Cleaner(Strathmore University, 2019-08) Odero, JerryData cleaning is an activity involving detecting and correcting errors and inconsistencies in a database, data warehouse or any data record of an organization. Kenya Revenue Authority (KRA) in its quest to be a fully data driven organization, is actively undertaking the data cleaning process. However, this process is currently manual and slow as it involves physical transfer of documents to be processed from the various stations, via different levels of management for approval, to the centralized return processing unit. A process, which might take at least a fortnight for the processing of one taxpayers ledger account. Furthermore, this whole process needs lots of man-hours, since there is a vast amount of data to be cleaned due to the many ledger accounts affected during the manual filing system that ended in 2014. There exists many data cleaning processes and approaches which are used to purge out dirty data, before it’s loaded into the data warehouse. These processes vary depending on the data source, they are time consuming and expensive for organizations, in terms of skilled staff and the tools involved, hence this research proposed the application of RPA (Robotic Process Automation) to develop an intelligent bot (Intellibot) to be used in the transactional data cleaning exercise in Kenya Revenue Authority (KM). With the transition from legacy system to I-Tax and I-CMS systems for domestic and customs revenue management respectively, the researcher sought to find out the current data cleaning process in the legacy system. This research led to the development of an RPA system for the current manual data cleaning process implemented and tested using the Blue Prism platform. The system detected the errors using a knowledge-based model-, clustering them as errors due to uncaptured returns, uncaptured losses or credit re-adjustments. The intellibot system was able to load the ledgers, detect the errors and clean them with utmost precision. Experiments conducted on performance of the bots varied by seconds, in the first experiment. Also in the second performance test, there was a variance of seconds in cleaning the different errors detected, hence improving the data integrity significantly: free of errors, to be migrated to the I-Tax platform, thus support better decision making process in the organization, and a higher return on investments.
- ItemInfluence of target-task approach of teaching on students achievement and retention in mensuration(Strathmore University, 2019-08) Ochihu, AmosThe purpose of the study is to compare the effectiveness of the use of target-task approach with conventional approach of teaching mensuration at senior secondary school level and determine which approach is more effective in teaching mensuration aspect of mathematics. It is also aimed at investigating whether any of the sexes improve more than other from the use of target-task approach. The study employed quasi-experimental design of non-randomized pre-test and post-test control group type. A sample of 248 SSI students was drawn from the population of 3842 SSI students in Oju local government area. Four research questions and four hypotheses guided the study. In each school, intact classes were used. Two set of lesson plans were developed for experimental and control groups respectively. Data were collected using mensuration achievement test (MAT). Research questions were answered using means and standard deviations while the hypotheses were tested using analysis of covariance (ANCOVA) at 0.05 level of significance. The finding among others shows that target-task approach was more effective in improving students’ achievement and retention than the conventional approach. It was also found that the use of target-task approach did not significantly differentiate between male and female students’ achievement and retention scores in mensuration. Based on the findings of this study, it was recommended that mathematics teachers should be encouraged to use targettask approach in their mathematics classroom among others.
- ItemInfluence of self-instruction on Mathematics achievement among students in secondary schools in Kenya(Strathmore University, 2019-08) Kati, RobertAchievement in Mathematics is a challenge to many students at secondary level in Kenya, for instance students in Vihiga Sub County. Despite the government’s effort in strengthening the subject, its performance is still wanting. The main objective of the study was to examine the influence of self-instruction on Mathematics achievement among students in secondary schools in Vihiga Sub-county. Self-instruction' was defined as a deliberate long-term learning project instigated, planned, and learned out by the learner alone, without teacher intervention. The Self- determination theory and Achievement Goal theory informed the study. The study adopted Mixed Methods approach and the Sequential Explanatory Design was used. The study targeted 1483 form four students, 35 Mathematics teachers, and 27 teacher counsellors. A sample size of 445 students, 11 Mathematics teachers, and 9 teacher counsellors were selected using stratified random, purposive and purposive sampling techniques respectively. Quantitative data was collected using a Students Questionnaire while qualitative data was collected using interviews from students, teacher counsellors and Mathematics teachers. Student’s achievement was assessed using K.C.S.E Exam results of the year 2017. Reliability of the questionnaire was ensured by Cronbachs alpha and a coefficient of alpha ¿0.7 was reported. Normality of data was tested by using Kolmogorov-Smirnov and Shapiro-Wilk (W) tests. Descriptive statistics such as frequencies and percentages were used to analyze quantitative data from questionnaires, while inferential statistics such as Regression Analysis and Pearson correlation coefficient were used to analyze quantitative data. On the other hand, thematic framework was used to analyze Qualitative data. The study found that there was statistically significant between self-instruction and Mathematics achievement(r =.192, n=396, p i.05). The findings showed that self-instruction predicted the achievement in Mathematics among secondary school students. The study recommended that the government in conjunction with the Ministry of Education should provide seminars and conferences for Mathematics teachers as a platform for constant reminder to teachers to avoid traditional modes of Mathematics teaching and embrace self-instruction strategy. This would enable Mathematic teachers to instruct their students in such a way as to enable them to take charge, control and evaluate their learning through self-instruction, hence enabling students to become autonomous learners in Mathematics. This is because the study reported that self-instruction has positive influence on Mathematics achievement among students in secondary schools.