Conferences/ Workshops/ Seminar proceedings and presentations

Permanent URI for this communityhttp://localhost:4000/handle/20.500.12092/1616

Welcome to conferences, workshops and seminar proceeding within Karatina University by Karatina University community

Browse

Search Results

Now showing 1 - 10 of 82
  • Thumbnail Image
    Item
    Integrating Artificial Intelligence Literacy in Library and Information Science Training in Kenyan Academic Institutions
    (SCECSCAL, 2024-09) Chepchirchir, Sally
    With the rapid technological advancements, Library and Information Science (LIS) programs should evolve to equip students in academic institutions with Artificial Intelligence (AI) skills and knowledge to meet the demands of the information profession. The objectives of this paper were to establish the current state of AI literacy in LIS training in academic institutions in Kenya, examine the extent to which AI literacy has been integrated into LIS curricula in academic institutions, identify the challenges and opportunities associated with the integration of AI literacy in academic institutions in Kenya, and propose critical recommendations that the management in academic institutions should consider for integrating AI literacy in LIS training in Kenya. The study employed a mixed-methods approach, combining qualitative and quantitative data collection methods. Quantitative data was collected through bibliometrics analysis, while qualitative data was collected using a systematic literature review and observation. Data was collected from Google Scholar using Harzing’s “Publish or Perish” software and academic institutional websites. It was analysed using Microsoft Excel, Notepad, and VOSviewer and presented using tables, graphs, and figures. The findings reveal that LIS professionals must possess essential skills and competencies in AI to meet the evolving needs of the job market. The study highlighted valuable practical insights and recommendations to the management in academic institutions on a comprehensive understanding of the opportunities and challenges presented by AI literacy in LIS training, offering a foundation for future research, policy development, and pedagogical innovation in the field.
  • Thumbnail Image
    Item
    Effectiveness of Reference Management Software in Enhancing Research Quality in Universities in Nairobi County, Kenya
    (SCECSAL, 2024-04) Kairigo, Samuel; Anduvare, Everlyn
    Reference management software (RMS) application is highly emphasised in academic research to improve research quality. However, studies raise concerns about their effectiveness since they have errors in functionality, language limitations, and inaccuracy of the citations and references generated. This study aimed to establish how effective RMS is in improving research quality. The study objectives were to investigate what reference management software is in use in the universities in Nairobi County, Kenya and to establish if reference management software programmes are effective in enhancing the quality of research in the selected universities. The study adopted a descriptive research design. This assisted the researchers in gathering data through a survey where an online questionnaire was administered to 18 respondents. All the universities within the County of Nairobi constituted the target population. The unit of analysis was the University Librarians because of their crucial role in promoting research quality in Kenyan Universities. In the analysis, insights were derived using a computer-based statistical package for social science (SPSS). The study revealed high satisfaction with RMS, particularly in terms of necessity, visual appeal, integration capabilities, and automatic formatting of references. While University librarians reported ease of navigation, there was limited awareness of alternative referencing approaches. Continuous institutional training programs for librarians and researchers on RMS usage, covering basic and advanced functionalities, are recommended.
  • Thumbnail Image
    Item
    Knowledge management structure and effectiveness of interventions on fall army worm (FAW) management among smallholder maize farmers in Kilungu, Makueni County, Kenya
    (RUFORUM, 2022) Aoko, P.; Chimoita, E. L.; Nzuve, M. F.; Maina, D.; Jadesola, E.; Syanda, J.
    Maize productivity in Kenya has reduced by 4.3 per cent, partly attributed to Fall Army Worm (FAW) infestation with yield losses of up to 37 % of the annual maize production over three years. Innovators have developed practices that manage FAW infestation and increase maize yield. Knowledge of these interventions is not available to all farmers and thus not applied. This occasioned for an investigation into the knowledge management and effectiveness of FAW control practices in maize producing regions of Kenya. This study conducted structured interviews with key informants and households in Kilungu, Makueni County. It sampled 387 respondents with a 95% confidence level and applied multi stage sampling. Statistical analysis using STATA found a significant influence of education and total income on the selection and adoption of FAW management practice. Handpicking recorded p values of 0.033 and 0.013 respectively. Analysis of adopted FAW management practices against maize output showed a significant effect from handpicking, use of pesticides, detergents and/or soil with p values of 0.099, 0.049, 0.025 and 0.075 respectively. Fellow farmers and workshops as sources showed a significant influence on maize output with p values of 0.012 and 0 respectively.
  • Thumbnail Image
    Item
    Effect of pre-harvest application of Chitosan and Silicon on growth, Lycopene content and shelf-life of tomato
    (2022) Gatahi, D.; Chimoita, E.; Kihurani, A.; Wanyika, H.
    Maize productivity in Kenya has reduced by 4.3 per cent, partly attributed to Fall Army Worm (FAW) infestation with yield losses of up to 37 % of the annual maize production over three years. Innovators have developed practices that manage FAW infestation and increase maize yield. Knowledge of these interventions is not available to all farmers and thus not applied. This occasioned for an investigation into the knowledge management and effectiveness of FAW control practices in maize producing regions of Kenya. This study conducted structured interviews with key informants and households in Kilungu, Makueni County. It sampled 387 respondents with a 95% confidence level and applied multi stage sampling. Statistical analysis using STATA found a significant influence of education and total income on the selection and adoption of FAW management practice. Handpicking recorded p values of 0.033 and 0.013 respectively. Analysis of adopted FAW management practices against maize output showed a significant effect from handpicking, use of pesticides, detergents and/or soil with p values of 0.099, 0.049, 0.025 and 0.075 respectively. Fellow farmers and workshops as sources showed a significant influence on maize output with p values of 0.012 and 0 respectively.
  • Thumbnail Image
    Item
    Developing an Early Warning System for Monitoring Drought and Ethnic Conflict for Poverty Alleviation in Tana River District, Kenya
    (African Institute for Health and Development, 2006-06-18) Amuyunzu-Nyamongo, Mary; Mwenzwa, Ezekiel Mbitha
    Internal and external conflicts have increased in African countries since independence and especially after the cold war. Whenever civil war occurs, it has often led to destruction of lives and property, leaving behind it a great trail of human suffering. In Kenya, where ethnic conflict has occurred in Rift Valley, Coast and North Eastern provinces, the impact has been devastating. Ethnic conflict in Tana River District can be traced back to 1948, which has resulted in the retardation of socio-economic development. It is against this background that the current study seeks to examine the timing, causes, consequences and the best practices to deal with drought and ethnic conflict. Specifically, it aims at developing an early warning system to monitor drought and ethnic conflict. The study will be carried out in the three divisions of Bura, Galole and Garsen in Tana River District, Kenya. This will be a collaborative study between the African Institute for Health & Development (AIHD), the government of Kenya and the communities. It will utilize participatory tools of data collection (historical timelines, seasonal calendar, problem analysis flow diagrams, focus group discussions and key informant interviews). An interviewer-based questionnaire will be administered to 685 respondents drawn from the three divisions. It is envisaged that minimizing the effects of drought and tensions that lead to ethnic conflict would significantly contribute to poverty reduction. Peace and security in this district would allow people to engage in agricultural and livestock production, and other income generating activities with minimal worries of possible attacks. The study is expected to contribute to the government’s and its development partners’ drought preparedness plans and conflict resolution not only in Tana River district but also in other areas with similar characteristics.
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
    Deep Transfer Learning Optimization Techniques for Medical Image Classification: A Review
    (2022) Kariuki, Paul Wahome; Gikunda, Patrick Kinyua; Wandeto, John Mwangi
    Medical image classification is a complex and challenging task due to the heterogeneous nature of medical data. Deep transfer learning has emerged as a promising technique for medical image classification, allowing the leveraging of knowledge from pre-trained models learned from large-scale datasets, resulting in improved performance with minimal training and overcoming the disadvantage of small data sets. This paper concisely overviews cutting-edge deep transfer learning optimization approaches for medical image classification. The study covers convolutional neural networks and transfer learning techniques, including relation-based, feature-based, parameter-based, and instance-based transfer learning. Classical classifiers such as Resnet, VGG, Alexnet, Googlenet, and Inception are examined, and their performance on medical image classification tasks is compared. The paper also discusses optimization techniques, such as batch normalization, regularization, and weight initialization, as well as data augmentation and kernel mathematical formulations. The study concludes by identifying challenges when using deep transfer learning for medical image classification and proposing potential future approaches for this field.
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
©Karatina University