Research Articles

Permanent URI for this collectionhttps://scholar.mzumbe.ac.tz/handle/123456789/364

Browse

Author: search.filters.author.Mbukwa, Justine N.End date: 2019

Search Results

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Statistical Analytic Tool for Dimensionalities in Big Data: The role of Hierarchical Cluster Analysis
    (Research India Publications, 2016) Mbukwa, Justine N.; Tabita, G Neeha; Anjaneyulu, GVSR; Rajasekharam, OV
    An interest for presenting this paper rose because of massive increase information with a very high dimensional from different sources in this era of globalization. Data are produced continuously and are unstructured (1). This paper is confined to literature review search for big data issue and challenges of several scopes in data. It brings a detailed discussion on the problem on these data and analysis done using the effective multivariate statistical tool namely clustering analysis technique as a data reduction technique. It is used as a base for discussion for existing challenge of multi-dimensionalities of data. The findings indicated that, the world is noisy due to massive flow of information continuously. Findings revealed that data emanating from face book, you tube and twitter can be used to predict the epidemic of influenza and even market trend (2 and 3). With face book data is used to predict the people`s interest. However, data from different sources have been proved to be useful in decision making efficiently and effectively for public as well as private sector. Cluster analysis technique sorts data/alike things into groups, to see if there a high natural degree association among members of the same group and low between members of different groups. Finally, this technique has proved failure to handle such heap of data with varied sources. With regards to data stored, it remains to be a challenge in terms of analysis among researchers and scientists. Therefore, it calls for advanced statistical software to cater for such an existing challenges.
  • Thumbnail Image
    Item
    Some Aspects of Correlation of Physical Capital and Infrastructures on Household Food Security: Evidence from Rural Tanzania
    (Journal of Economics and Sustainable Development, 2014) Mbukwa, Justine N.
    To achieve the first Millennium Development Goals is still a challenge. The problem of poverty in the context of hunger still persists in Tanzania. Household’s members have not sustainable access to enough and quality of food. The major objective of this study is to ascertain whether exists some aspects of correlation between physical capital and infrastructures on the households` food security. This study was carried out in rural part of Tanzania, in Mvomero district covering three villages selected randomly (a total sample 0f 382 households). Data analysis was done using SPSS version 15.0. Chi-square test was adopted for plausible analysis assesses the extent to which some correlation exists between food security status of the households against independent variables (physical capital and infrastructures of the households). 2 χ = 6.963; − p value Based on the data analyzed empirically, it is remarkable by evidence that variables such as pesticides ( = ( 2 χ = 13.343; − p value = food security in the study area. 0.008 ), tractors ( 2 χ = 0.000 10.024; − p value = 0.002 ) and electricity ) were found to be statistically significant correlated with household’s In view of these findings, there is a need to pay attention supporting rural farmers’ to be able to access farm inputs because of existing some correlation with the household food security status. Finally, this study recommends further study to be carried by incorporating advanced statistical model such cluster analysis, principal components and factor analysis which deals with large data for plausible and interpretable findings.
  • Thumbnail Image
    Item
    A model for predicting food security status among households in developing countries
    (International Journal of Development and Sustainability, 2013) Mbukwa, Justine N.
    Food security prediction has been challenging aspects in developing countries particularly in African countries such as Tanzania. Consequently, government lack proper stimulated information that is necessary in making decision on efforts required for stabilizing food situation and status in their countries. Scientifically it has been observed in research and practical that this is caused by lack of proper mechanisms, tools and approach suitable for modeling and predicting food status among households. This paper proposes a logistic regression based model for analysis and prediction of food security status. The proposed model is empirically test using practical data collected from one district in Tanzania.