Research Publications
Permanent URI for this communityhttp://localhost:4000/handle/20.500.12092/1797
Welcome to Karatina University Research papers. The community has research papers produced by Karatina University Community
Browse
9 results
Search Results
Item Evaluating the Performance of Tree-Based Predictive Models as Programme Recommenders for University Entrants in Kenya.(2024-10) Kabiru, Kibuthi J.; Makiya, Ratemo C.; Anduvare, E. M.Enrolling for the wrong programme by university students has, to an extent, contributed to the high rates of discontinuation on academic grounds, repeat year cases, change of programme after registration, interuniversity transfers, deferments to change programme, drop out cases, suspension over exam irregularities as well as to strikes. This study focused on finding a technological solution for reducing these cases by evaluating three tree-based predictive models and recommending the most predictive model to implement as a programme recommender. Data was collected in five selected public universities in Kenya using Google Forms. The respondents were 308 translating to 308 rows of data with 36 columns. Numpy, Pandas, Matplotlib, Sklearn, Seaborn, Scipy, Plotly python analytics libraries were deployed using Jupyter Notebook for Anaconda. The cleaned and processed dataset features had categorical variables thus one-hot-encoding technique was employed. Data was split for training and testing with the random_state set to 42. Gini index criteria was implemented. The three models were evaluated on their performance from the optimally split data for training and test with a 80:20 ratio. Random Forest (RF) came out the most predictive at 99.3% followed by Gradient Boosting (XG Boost) at 90% then Decision Tree (DT) at 80.93%. The testing accuracy score for RF was 81.72%, XGBoost was at 75.72% and DT was at 76.34%. Confusion matrix criterion was implemented to evaluate the performance of the three models. The results of this study have demonstrated the high accuracy level of RF as the most predictive tree-based model for this real-world University crisis. The model is recommended for development as a system to be integrated into the KUCCPS portal. The integrated system is dubbed Programme Recommender which if launched would highly predict the best programme of study for application by university entrants.Item Causal Factors Responsible for Changes in the Attributes of Urban Green Spaces in Nairobi City County, Kenya(2024-07-26) Mbugua, Martin Wandie; Kamiri, Hellen; Kamau, PeterUrban green spaces are an important part of public open spaces and a common service provided by a city, town, or municipal council. In Nairobi City County, green spaces have been increasingly threatened by overcrowding, poor planning, weak management structures, and illegal alienation thus denying city residents access to the much-needed recreation and leisure facilities. As the population increases in urban areas, their activities impact the environment and therefore the ecosystem services. This study was done with the Principle of Intelligent Urbanism in mind, as the most prolific urban development theory developed by the International Congress of Modern Architecture. Quantitative and qualitative methods were used to collect data while a cross-sectional survey designs and stratified random sampling of the green spaces was employed, based on the location within the urban core and peri-urban of Nairobi County. Four green spaces (Karura Forest, Ngong Road Forest, Nairobi Arboretum, and City Park) with diverse characteristics and attributes were selected. A sample population of 384 visitors to the green spaces was surveyed for their perceptions of the functions, uses, and benefits of the green spaces. Our findings showed that management structures and population growth are the main causal factors responsible for changes in the attributes of urban green spaces.Item Temporal changes in Rainfall and Temperature influence on Solid Waste Generation in Bungoma County Urban Centres, Kenya(2024-07-28) Wafula, Godfrey Wekesa; Ouna, Tom; Kamiri, HellenGlobally, temperature and rainfall play an integral role in the management of solid waste in urban areas. This is because many countries especially, the developing countries conventionally rely on both variables in the management of waste. However, the urban environment faces many threats as a result of challenged waste management practices due to these climate stressors. This research aimed to assess how temporal changes in temperature and rainfall influence solid waste generation. The study was carried out in Bungoma county urban centres, specifically Bungoma town, Webuye, Kimilili, Kapsokwony, Chwele, and Sirisia townships. The study applied a stratified sampling technique to select the urban and peri-urban centres while random sampling was used to select respondents in residential dwellings, business people, urban residents, and officers from government institutions including public health and environment, water, and natural resources were interviewed. Ancillary data was collected using questionnaires and interview schedules while historical data was obtained to validate observed data. Data was analysed using descriptive and inferential statistics and summarized using tables, charts, and graphs. The findings of the study indicated that rainfall and temperature significantly influenced the generation of solid waste.Item Assessment of Solid waste generation and composition in Bungoma County Urban Centres for appropriate waste management approaches(2024-08-02) Wafula, Godfrey Wekesa; Kamiri, Hellen; Ouna, TomThe quantity and characteristics of urban solid waste are a result of the interconnectedness of diverse factors at play including social, economic, and environmental factors. These directly influence the choice of management practice at the household or community level. Understanding these relationships can help policymakers and urban planners develop more effective waste management strategies. This study aimed at assessing solid waste generation and composition in Bungoma county urban centers to inform on strategies for effective management of urban waste. The study was carried out in Bungoma County urban centers, specifically Bungoma town, Webuye, Kimilili, Kapsokwony, Chwele, and Sirisia townships. The study applied a stratified sampling technique to select the urban and peri-urban centers while random sampling was used to select respondents in residential dwellings, business people, urban residents, and officers from government institutions including public health and environment, water, and natural resources were interviewed. Ancillary data was collected using questionnaires and interview schedules while historical data was obtained from the Bungoma County Department of Environment and used to validate observed data. Data was analysed using descriptive and inferential statistics and summarized using tables, charts, and graphs. Our findings show that the highest waste was generated from residential homes, commercial, and municipal services. In terms of composition, biodegradable waste had the highest proportion, followed by plastics and glass waste. Population growth could be a key factor in the increased generation of the waste.Item Attributes of Urban Greenspaces and their Influence on Visitors’ Preferences in Nairobi City County, Kenya(Eastern African Journal of Humanities and Social Sciences Vol. 3 No. 1, 2024-07-07) Mbugua, Martin Wandie; Kamiri, Hellen; Kamau, PeterUrban green spaces refer to land covered with vegetation such as forests, street trees, parks, gardens, and water bodies in an urban setup. In this study, we sought to investigate the attributes of urban green spaces in Nairobi County and their relationships to visitor preferences. The Garden City Model advanced by Howard (1898) guided the study. Quantitative and qualitative methods were used to collect data while a cross sectional survey designs and stratified random sampling of the green spaces was employed, based on the location within the urban core and peri-urban of Nairobi County. Four green spaces (Karura Forest, Ngong Road Forest, Nairobi Arboretum, and City Park) with diverse characteristics and attributes was selected. A sample population of 384 visitors to the green spaces were surveyed for their perceptions of the functions, use, and benefits of the green spaces. Our findings showed that attributes of green spaces including location, accessibility, security, hygiene, and infrastructure could explain the interaction between green space provision factors and the frequency of visits, time spent in the green spaces and overall satisfaction of the visitors. Green space provision should be equitable in regard to distance to residents, quality of spaces, facilities and services and should be designed to meet the needs of diverse residents. Further, they should provide services and benefits such as shade, recreation, and health which are the main attractions to the green spaces. To achieve maximum benefits for visitors, green spaces need to be safe and physically accessible to all.Item Management Strategies of Prosopis juliflora in Eastern Africa: What Works Where?(2024-04-22) Kamiri, Hellen Wangechi 1,2, *; Choge, Simon K. 3; Becker, Mathias 2Abstract: Prosopis juliflora is a shrub/tree originating from the Americas. Since its introduction for fuel wood afforestation into eastern Africa, it has been invading crop- and rangelands with negative effects on the environment and on livelihoods. Understanding the management strategies for Prosopis and matching them with ecological, social-cultural, and economic needs of the eastern African region is a pressing scientific issue. We analyzed management strategies of Prosopis, focusing on determinants and drivers of their choice of applied management strategies as well as their effectiveness. We identified 1917 scientific contributions published between 1970 and 2022. Following a multi-step screening, we reduced the references to 53 relevant (internationally) published papers with a focus on the management of Prosopis in the east African region. Analysis of the literature shows that factors driving invasion dynamics but also land users’ social-economic as well as cultural attributes determine the type of management strategy and shape local control actions. Main strategies comprise (1) physical containment of invasive spread, (2) chemical, mechanical and biological approaches to reduce stand densities, (3) complete eradication, (4) restoration of invaded land, and (5) economic use of Prosopis products. Adopted strategies are based on actual and perceived impacts of invasion, and the adoption and success of individual strategies is highly location specific.Item Salinity tolerance, growth and survival of three Artemia franciscana (Kellogg, 1906) populations under laboratory conditions(Wiley, 2024-04) Ngarari, Morine M.; Rugendo, Derrick G.; Nyonje, Betty M.; Hinzano, Sheban M.; Midumbi, David O.; Ngugi, Charles C.; Gatune, Charles W.In the 1980s, Artemia franciscana from San Francisco Bay (SFB) was introduced into Kenyan saltworks, where it has colonized and established stable populations. However, little is known about its biology, particularly with respect to its parental SFB population. This study compared the salinity tolerances of Kenyan (KEN) population, their SFB progenitors and those of Great Salt Lake (GSL) populations. Growth and survival of these A. franciscana populations were evaluated under varying salinity levels in a laboratory set up. A. franciscana nauplii were cultured at a rate of 1 nauplii/mL in 36 Erlenmeyer flasks and fed microalgae (Chaetoceros sp.) at 1.5 × 106 cells/animal/day for 8 days. Survival was evaluated daily and survivors were fixed in individual vials with Lugol solution. The total length of each fixed A. franciscana nauplii specimen was measured under a compound microscope. All populations were susceptible to salinities greater than 100 g/L. Compared with the parental SFB population, the KEN population exhibited significantly reduced survival and growth at 140 g/L, suggesting a narrower salinity tolerance range. These findings underscore the need for further studies focusing on other physiological parameters, abiotic factors and genetic characterization to confirm whether the KEN population is experiencing ecological adaptation. This will contribute to the optimization of Artemia practices in various salinity environments as a result of climate change.Item Salinity tolerance, growth and survival of three Artemia franciscana (Kellogg, 1906) populations under laboratory conditions(Wiley, 2024-04-17) Ngarari, Morine M.; Rugendo, Derrick G.; Nyonje, Betty M.; Hinzano, Sheban M.; Midumbi, David O.; Ngugi, Charles C.; Opiyo, Mary A.; Okalo, Francis A.; Gatune, Charles W.In the 1980s, Artemia franciscana from San Francisco Bay (SFB) was introduced into Kenyan saltworks, where it has colonized and established stable populations. However, little is known about its biology, particularly with respect to its parental SFB population. This study compared the salinity tolerances of Kenyan (KEN) population, their SFB pro-genitors and those of Great Salt Lake (GSL) populations. Growth and survival of these A. franciscana populations were evaluated under varying salinity levels in a laboratory set up. A. franciscana nauplii were cultured at a rate of 1 nauplii/mL in 36 Erlenmeyer flasks and fed microalgae (Chaetoceros sp.) at 1.5 × 106 cells/animal/day for 8 days. Survival was evaluated daily and survivors were fixed in individual vials with Lugol solution. The total length of each fixed A. franciscana nauplii specimen was measured under a compound microscope. All populations were susceptible to salinities greater than 100 g/L. Compared with the parental SFB population, the KEN population exhibited significantly reduced survival and growth at 140 g/L, suggesting a narrower salinity tolerance range. These findings underscore the need for further studies focusing on other physiological parameters, abiotic factors and genetic characterization to confirm whether the KEN population is experiencing ecological adaptation. This will contribute to the optimization of Artemia practices in various salinity environments as a result of climate change.Item Macroinvertebrate metrics and lipid profiles as potential indicators of land use influence in a high altitude tropical highland stream (Sagana River Basin, Kenya)(Elsevier, 2024) Odhiambo, Charles O.; Ngugi, Charles C.; Sharma, Rekha R.; Oyoo-Okoth, ElijahLand use practices near river ecotones result in numerous inputs of terrestrial materials into the river ecosystem. While macroinvertebrates population metrics is widely used in monitoring water quality changes, including those induced by humans, the use of lipid profiles in these organisms to monitor influence of land use practices is rather ignored. The aim of this study was to determine lipid profile in macroinvertebrates as potential indicator of human influence in Sagana River Basin, a high altitude tropical highland stream in Kenya. Sites were selected based on differences in land use activities delineated: forest with minimal land use activities, while cropland and saw milling areas had varying degrees of land use activities, with urban areas designated as sites with most human impacts. The macroinvertebrate metrics such as occurrence, abundance, diversity, percentage of oligo chaetes and chironomids (%OC), as well as the Ephemeroptera, Plecoptera and Trichoptera (%EPT) index differed relative to land use changes. Variation in the Fatty Acids (FAs), alcohols and sterols in benthic mac roinvertebrates was related to the land use activities along the stream. Evidently, streams traversing the urban areas had higher concentrations of Phospholipid fatty acids (PLFA), while streams traversing croplands had the highest neutral lipid fatty acids (NLFA). The dominance of monounsaturated FAs (MUFAs) in forest, shortchain FAs (SCFAs) and polyunsaturated FAs (PUFAs) in croplands and longchain FAs (LCFAs) near urban areas clear indicate different sources of these materials, mainly from terrestrial sources. Individual fatty acids, alcohols and sterols profiles discerned difference in land use changes. The concentrations of SCFAs, MUFAs, LCFAs, TeoH, 27Δ5.22, 28Δ5,24, 28Δ5,22, 29Δ5,22 and 29Δ5,22 in the benthic macroinvertebrates samples were positively associated with OC%, EPT, total abundance and eH’. Meanwhile the concentration of phytol, AqOH, 27Δ5 and SCOH were associated with Simpson index. The current findings support the combination of macroinvertebrates species composition, abundance, diversity and lipids profiles to detect land use activities at the riverine scale. While implementing catchment-based river management and conservation activities should incorporate lipid analysis as a management tool. There is need for continuous monitoring of long term trends in land use activities on the changes in lipid content of the macroinvertebrates