MSIS Theses and Dissertations (2017)

Permanent URI for this collection

http://hdl.handle.net/11071/5602

Browse

Now showing 1 - 9 of 9

An Algorithm for predicting road accidents based on traffic offence data
(Strathmore University, 2017) Jwan, Levice Obongo
Drivers with multiple records of road traffic violations for instance speeding, driving under influence of alcohol and using mobile phones while driving have been considered as a high risk group for possible involvement in road accidents. Studies have shown that there are links between these reckless behaviors and road accidents. It is therefore critical that such drivers be identified early in advance to eliminate that likelihood. Currently, the road traffic offence data collected by National Transport and Safety Authority for instance speeding and drunk driving data is solely used for reporting and prosecution hence not adequately utilized in ensuring road safety. Effective utilization of these data can positively impact road safety management since authorities can put in place mitigation mechanisms in order to prevent the frequent road accidents. The algorithm-based system developed in this study makes use of traffic offence data to predict the likelihood of a driver causing road accident. Data was gathered using close-ended questionnaires and interviews. The questionnaires and interviews intended to determine causes of road accidents and specific aspects about; booking an offender, relaying of traffic accident data and the need for a system among users within the transport sector in Kenya. Three categories of respondents were used; the National Transport and Safety Authority, the Kenya Police and the motorists. Similar questionnaires were given to the police and the NTSA officials while the motorists had their own set of questions. From the research, it emerged that the major causes of accidents in Kenya were; speeding, dangerous overlapping and drunk driving. Of the 37 respondents; 22 supported the algorithm-based system, indicating a 59.47% approval for the system. The implication of the research is that there will be more people booked for traffic offences and it will be possible for law enforcement to know the risk level of a driver based on the offences committed.
A Healthy nutrition model for infants
(Strathmore University, 2017) Machira, Hope Njeri
Accessing information on nutrition for a majority of young mothers in the sub- Saharan region has been marked with numerous challenges. The most common method for gathering nutrition information is by searching for the information on the internet. This process of gathering information does not always result in providing the needed information and mothers also find the information unclear in many occasions. As such, this has resulted in widespread cases of under-nutrition in infants due to poor feeding practices. The current technological platforms for disseminating nutrition education fail to focus on the unique nutritional needs of an infant. This research aimed at understanding the challenges faced by mothers in meeting their infant’s nutritional requirements, to understand the methods and tools used to provide nutrition literacy and establish the data and information requirements for determining the nutritional value of foods for infants. The ability to feed healthy food begins with awareness of the nutritional value in food. A healthy diet is most important during infancy than during any other time of life because it impacts on brain growth, development of the nervous system, entire growth and development of the body and for eradication of diseases such as rickets and anemia caused by lack of essential nutrients. The study proposes the use of a nutrition model for infants for generating healthy meals according to their age. The proposed prototype made use of the guidelines as per the Recommended Dietary Allowance (RDA) which is provided by the Institute of Medicine (IOM) and using rules determined the daily meals that would provide the required nutrients an infant’s needs. Agile methodology was used to develop the prototype, it allowed for a quicker release and subsequent user feedback. The prototype can be accessed through internet browser both on mobile phone and on a computer. The model was tested through a simulation of various modules. Of the 10 participants that took part in this exercise, 6 were satisfied that the model fulfilled the intended functions of generating healthy meals for infants. 7 of the respondents also found the model easy to use. The model was therefore successful in offering nutrition expertise for infants.
Mobile application framework for collaborative creation and sharing of Jua Kali product designs: a case study of Kamukunji Enterprise Jua Kali Cluster
(Strathmore University, 2017) Gikera, James Njihia
The Informal jua kali sector is a vibrant entity that has created employment and innovative products in Kenya. Many homes in Kenya have at least one or more jua kali products. Over time, some products have evolved through incremental improvements to become award winning products. The jiko for instance has evolved into a number of fuel-conserving variants, some of which have garnered more merit than others. For instance, the Jikokoa and Jiko poa are highly polished ceramic types that have gained local and international acclaim. Both are patented trademarks produced by Burn Design Laboratories. However, not all jua kali products gain this kind of traction. The result is that many product designs are phased out naturally by demand and supply. The manual storage of product designs and sketches on paper is not reliable as they wear out, get lost or are stolen over time. Some artisans even prefer not to have any physical records of their designs for fear of duplication by others. This study therefore sought to propose the creation of a mobile application framework for collaborative creation and sharing of jua kali product designs to be used among artisans. The application allows direct sketches to be made on a smart device’s touch screen and upload of images while helping track and attribute each contribution to the relevant artisan. This, the researcher hopes, helps in creating more customer-centric and uniform products with higher quality and less competing variations. The application has a user-interface and was designed as a web application to run on any internet enabled device with a web browser. The study utilized a mixed research methodology with both qualitative and quantitative data collection techniques. This enabled the researcher to understand the challenges and methods used and to analyze collected data in order to come up with the framework based on the results. 72% of the respondents indicated they would use the application to manage and store designs and sketches but only 36% would use it for collaborative creation citing lack of trust and exposure to risk as the main hindrance. This informed the researcher to create functionality for either doing private deigns or collaborating and sharing designs in a public domain. The objectives of this study were to identify the challenges associated with the current methods, review the existing online platforms, to develop a mobile application framework and to validate the mobile application framework for collaborative creation and sharing of jua kali products.
Near field communication based-model for authentication in online banking
(Strathmore University, 2017) Omondi, Esther Akinyi
Online banking has enabled bank customers to perform their banking activities at the comfort of their homes as opposed to physically visiting the bank branches. In the banking environment, authentication is very crucial because the bank should be able to give access to the only authorized customers. Currently, there are various authentication methods available to the banks for authenticating their customers. However, the empirical study has shown that there is an increasing number of identity theft leading to huge financial loses for both banks and bank customers. Additionally, the better authentication systems are complex for customers and more costly to the banks. This dissertation discusses the use of Near Field Communication (NFC), a short range based wireless communication technology currently improving the consumers’ usability due to its ability to exchange digital material as well as connecting electronic devices remotely. It is an evolving technology employing the use of Radio frequency identification (RFID) that lets electronic devices like Smart phones interconnect over very close range. The NFC technology has been integrated into some available Smart phones, when employed together with other authentication mechanisms such as Master Card’s Chip authentication program (CAP), its usability level is increased. The model employs the use of NFC enabled Smart phones and NFC enabled bank cards as third factor authentication. Agile methodology was used for the model development, and a java code that generates a QR code was developed. The NFC based model, when implemented is able to eliminate the need for a hard token which is an extra baggage to the customer and additional cost to the bank. Consequently, the NFC technology enhances security for online banking by protecting against online identity theft as well as form basis for future research in NFC application the banking industry.
A Prototype for predicting real estate investment performance in Kenya
(Strathmore University, 2017) Kihumba, Moses Kimani
Predicting investment performance is central to attracting investors in any property or business venture. Investors are keen to predict the future in order to protect their investments and choose assets with the best returns. All asset returns are predictable to some extent, with returns on real estate relatively easier to forecast due to the nature of assets. Forecasting is thus an important component in property investment decision-making. Currently, majority of investors in Kenyan real estate sector, rely on speculation and sales comparison to make investment decisions. Multiple regression models have been applied successfully in forecasting real estate investments in other markets. They incorporate socio economic variables, housing and proximity characteristics to estimate the value of real estate assets. The researcher applied a multiple regression model for predicting house prices by setting house price as the dependent variable (Y) while holding the Gross Domestic Product, income of households, cost of land and housing units developed as the predictor variables (X).This predicted house prices (Y) on the basis of the X variables and determined the influence of the variables on the price. Agile development methodology was applied in the development a web application that integrated the forecasting model, an analytical backend helps to present the forecasts to investors in terms of figures, charts, and graphs that are easy to interpret and compare. Various tests were also performed on the prototype including integration and system tests. User acceptance testing was also carried out where majority of the respondents found the interface easy to use, and indicated that the application met its stated objectives as outlined in the usability questionnaire.
A Recommender system for rental properties
(Strathmore University, 2017) Guy, Nicolas Nahimana
Locating products or services online that meet users’ needs is increasingly difficult due to the large pool of choices to consider before arriving at the desired one. A user may spend a considerable amount of time exploring numerous online resources to locate items that fit his requirements. Furthermore, users may not always express their preferences in a manner that easily matches them to items that could meet them. Searching for items online has been done mainly through database queries that return a list of the most suitable items. Recommender systems technology can be applied to ease the task of locating desired items online. This study proposes a recommender system that enables users to carry out a preference-based search on rental properties and enables them to refine those preferences using example-critiquing in case they are not satisfied with initial search results. This recommendation approach has been shown to provide more accurate search results. This research adopted the Object-Oriented Systems Analysis and Design (OOAD) approach to the development of the system. The system was developed as a Web application using the Ruby on Rails framework. Furthermore, the system was tested to ascertain that it performed as designed.
RFID-Based model for remote chicken monitoring: case of free-range chicken farming in Machakos County
(Strathmore University, 2017) Nthusi, Gift Musyoka
Poultry farming is an important economic activity among Kenyan farmers. In order to earn income, they have to raise chicken and sell them once fully matured. This Indigenous chicken farming not only requires time investment but also money. However, security of chicken is an inevitability every farmer has to face. This is because, this category of chicken needs a free-range environment where it can move around, scratching for food, worms and other edibles. The chicken farmer faces challenges majorly theft of chicken by workers or strangers who randomly visit the farm. In this study, we developed a model that tracks chicken using RFID tags. This model enables a farmer to remotely monitor chicken from anywhere using the internet. The model constantly records and stores chicken data captured by the reader and updates this information in a database. The farmer can thereafter retrieve the tagged chicken information using a user interface designed to simplify the process of monitoring the chicken to minimise theft. The RFID model was designed to identify the chicken, each with its specific tag, containing particular information about it. To demonstrate this, an Arduino Uno microcontroller is set up with an encoded RFID system for chicken tracking. The RFID system uploads data sketches to a remote server via an Ethernet shield.
A Scheme for improving data confidentiality in the cloud computing environment
(Strathmore University, 2017) Tikwang, Iriale Fabiano
Cloud computing has ushered in an era whereby small and medium sized companies enjoy computing power which was usually a preserve for big corporations. Despite these benefits, the present cloud data confidentiality techniques are still evolving, and as they evolve so are the threats, hence posing security and privacy challenges, thus becoming an impediment to cloud adoption. Currently, cases have been cited where hackers have stolen stored cloud data, later to appear in social media embarrassing the firms. Among the key vulnerabilities attributed to loss of cloud data include: account hijacking, malicious insider breaches, data breaches attributed to weak Identity and access management, phishing, SQL injection, among others. Several research articles have been reviewed with some proposed solutions but these solutions have fallen short of addressing account hijacking and malicious insider threats. In addition, the online survey conducted highlighted that insider breaches are among the main form of vulnerability to cloud data. These challenges within the cloud storage informed the basis for the design of a scheme for improving data confidentiality in the cloud computing environment. The data confidentiality is achieved by implementing authentication login which triggers a six digit code to be sent to a client mobile or e-mail for further authentication, thus, enabling situational awareness of data breaches in real-time. This approach will enhance reliability and trust of cloud services enabling users to maximize on potential benefits offered by the cloud environment.
A Tool for predicting the likelihood of information systems implementation failure: case of Kenya National Highways Authority
(Strathmore University, 2017) Masita, William Okari
Every implementer of information systems would desire to have a successful project implementation. To achieve this, they work towards eliminating causes of failure. Therefore, there is need to know these causes before implementation begins. This study sought to investigate the causes of information systems implementation failure and come up with a tool to predict likelihood of failure. To achieve this objective, relevant literature was reviewed covering approaches used in implementing information systems, ways of measuring success in implementation and techniques used in mitigating failure. Both quantitative and qualitative research design methods were used to study information systems implementation projects at Kenya National Highways Authority (KeNHA). Arising from the findings, it was found that IS implementation projects fail due to inadequate user involvement, changing scope, requirements not well understood, poor estimation techniques, failure to manage expectations of users, failure to manage the implementation plan, failure to adapt to business change, lack of commitment to a systems implementation methodology, poor user commitment, lack of top management support, insufficient staffing, team members lack of requisite knowledge and skills and inadequate people management skills. In order to come up with the proposed solution , the research findings were analysed through use case diagrams and sequence diagrams. Following this analysis, the system was designed using design class diagrams and entity relationship diagrams. Using this design, the system was built using PHP 5.6. MySQL database was also incorporated at the back end. The system incorporates main causes of failure in implementation of information systems. Existence of these factors in any information system implementation project is analysed taking into consideration their relative weights and materiality. Once analysed, a report will be produced to indicate whether there is likelihood of failure. This tool is advantageous in that there will be early warning in case failure is likely and therefore organisations will institute corrective mechanisms prior to commencement of the implementation process.

Browse

Browsing MSIS Theses and Dissertations (2017) by Title

Results Per Page

Sort Options