SIMC 2019
Permanent URI for this collection
Browse
Browsing SIMC 2019 by Title
Now showing 1 - 20 of 99
Results Per Page
Sort Options
- ItemA 2D-approach towards the detection of distress using Fuzzy K-Nearest Neighbor(Strathmore University, 2019-08) Machanje, Daniel; Orero, JosephThis research focuses on a novel approach of distress detection referred to as the 2D approach, using the fuzzy K-NN classification model. Unlike the traditional approach where single emotions were qualified to depict distress such as fear, anxiety, or anger, the 2D approach introduces two phases of classification, with the first one checking the speech excitement level, otherwise referred to as arousal in previous researches, and the second one checking the speech polarity (negative or positive). Speech features are obtained from the Berlin Database of Emotional Studies (BDES), and feature selection done using the forward selection (FS) method. Attaining a distress detection accuracy of 86.64% using fuzzy K-NN, the proposed 2D approach shows promise in enhancing the detection of emotional states having at least two emotions that could qualify the emotion in question based on their original descriptions just as distress can be either one or many of a number of emotions. Application areas for distress detection include health and security for hostage scenario detection and faster medical response respectively
- Item3-Group divisible designs with block size 4 and 2 groups(Strathmore University, 2019-08) Bezire, WilbroadOne of the main problems in design theory is to find construction methods to obtain unknown designs. In this paper we give fundamental constructions of group divisible SBS designs with block size 4 and 2 groups. We give necessary conditions for their existence and prove that these conditions are sufficient. We use tools from graph theory and finite fields, for example-designs, Steiner quadruple systems (SQSs), 1 and 2-factorisations of complete graphs.
- ItemAccelerated circularity of consumer electronics using Block Chain and Optimized Machine Learning Algorithms(Strathmore University, 2019-08) Mwongela, StanslausAt the heart of our economy lies a linear take, make, dispose model, relying on consuming big amounts of finite products and fossil fuels. The challenge of consumer electronic waste is colossal, and is continuously growing. In 2016 alone, there was a global generation of 44.7 million tons of e-waste, with an estimated raw material value of EUR 55 billion. 435 thousand tonnes of the 44.7 million tonnes were mobile phones. Only 20 percent of e-waste is documented to be collected and recycled under appropriate conditions, whereas the remaining 80 percent is either thrown into the waste stream or dumped, traded or treated under substandard conditions. The consequences of this linear electronics system pose environmental and health impacts both at the production and disposal ends. At the same time, the economic opportunity is significant. Taking smartphones as an example, almost 1.5 billion are shipped every year, with each unit containing components worth over USD 100. This represents a potential USD 150 billion of value that enters the market each year. This value should remain in the system. Even if the materials present in smartphones were recovered through recycling the least valuable loop of a circular economy they could be worth up to USD 11.5 billion. The need to rethink the system is clear, but solutions so far have been limited. The circular economy coupled with the Al opportunity presents a vision and a workable path forwards to meet these challenges, by redesigning our economy to be one that creates rather than extracts value, and keeps the finite technical resources in flow within the economy. The main purpose of this study is to build a Block Chain Circularity Incentive System that facilitates the accelerated circularity of consumer electronics. The main focus being a decision system that processes the block chain data of consumer electronics use and disposal needs to allow circular product and material flows as well as operate circular business models. A formal experimental design is proposed, more specifically, a simple factorial design. As opposed to having one learning algorithm to process electronic products usage on a Big Chain DB, this research proposes to run multiple algorithms and aggregate the results into a single score. The research also proposes to collect primary data from consumer electronics manufacturers, consumers of electronics, and E-waste recovery firms. The primary data would be used to identify the challenges this stake holders face, train the learning algorithms based on consumer electronics use and disposal needs and additionally test the Block Chain Circularity Incentive System developed.
- ItemAIDS-related Kaposis sarcoma pathogenesis(Strathmore University, 2019-08) Chimbola, Obias; Lungu, Edward M.Kaposis sarcoma (KS), the most common tumor associated with human immunodeficiency virus- I (HIV-I) and human herpesvirus-8 (HHV-8) (also referred to as Kaposis associated herpesvirus (KSHV) infection, develops in approximately 20 percent of patients infected with HIV-I. The lesions of this multicentric vascular neoplasm are purplish patches, plaques, or nodules. Although highly active antiretroviral therapy (HAART) can prolong the time to treatment failure in patients with KS, nearly every patient with AIDS-related (or epidemic) KS eventually develops disseminated disease. Progression usually occurs in an orderly fashion from new localized or widespread mucocutaneous lesions to more numerous lesions and generalized skin disease with involvement of lyrnph nodes, gastrointestinal tract (GIT), lungs and other organs. Studies have shown that KS cells themselves are not infected with HIV-I; therefore, it is widely accepted than HIV-I does not play a direct oncogenic role in AIDS-KS. However, the precise role of HIV-I in AIDS-KS is still not completely understood, and there is considerable debate over whether HIVI plays a passive role (through the induction of immunosuppression) or a more direct role in the pathogenesis of this disease. We formulate a mathematical model to study the dynamics of HIVI related KS pathogenesis. In this model, it is assumed that HIV-I infects only the CD4 + T cells and HHV-8 infects the B- cells, which largely remain latently infected and only become reactivated after exposure to inflammatory cytokines and other growth factors secreted from HIV-I infected cells. The Infection free and Infection persistent equilibria have been found and their stability established. It is found that the disease can exist even if both sub-group reproduction numbers, relating to HIV-I and HHV-8, are less than unity.
- ItemAnti-N-Order Polynomial Daugavet Property on Banach Spaces(Strathmore University, 2019-08) Emenyu, JohnWe generalize the notion of the anti-Daugavet property (a-DP) to the anti-N-order Almasi room, polynomial Daugavet property (a-NPDP) for Banach spaces. The characterization SBS of the a-NPDP is through the spectral information; however, it is well-known in nonlinear theory that there is no suitable notion of the spectra for nonlinear operators resulting into enormous structural challenges to the known characterization techniques for the a-DP. To bypass some of the problems, we establish that a good spectrum of a nonlinear operator is one whose associated eigenvectors are of unit norm and study the a-NPDP for locally uniformly convex or smooth Banach spaces (luacs); in particular, we prove that locally convex or smooth finite dimensional Banach spaces have the a-mDP for rank-I polynomials and then extend this result to innite dimensional luacs Banach spaces. Besides, we prove that locally uniformly convex Banach spaces have the a-NPDP for compact polynomials if and only if their norms are eigenvalues, and moreover, uniformly convex Banach spaces have the a-NPDP for continuous polynomials if and only if their norms belong to the approximate spectra. As a consequence of these results, we conclude that all continuous In-homogeneous polynomials that satisfy the N-order polynomial Daugavet equation on a uniformly convex Banach space such as Lr-spaces for 1 < r < 1 and Hilbert spaces have nontrivial invariant subspaces; this result was not known.
- ItemApplied arithmetic geometry(Strathmore University, 2019-08-12) Dr. Ambrus, PalThe aim of this lecture series is to introduce some methods of arithmetic geometry which are applied in cryptographic research. Cryptography, including more sophisticated versions such as elliptic curve cryptography, allows for efficient protocols for information security, and is widely used in the banking sector including mobile money transfers, an industry in which Africa is a world leader. The methods presented can be used by African research groups to tackle a range of problems arising in technological challenges relevant to the African development context. Arithmetic geometry is a rather modern, highly prestigious and very developed area of pure mathematics, developed originally for studying Diophantine equations. I has very efficient methods to count points on algebraic varieties over finite fields which is closely related to the original motivating problem of finding rational points on varieties over number fields, a geometric reformulation of Diophantine equations. The former problem is very important in cryptography and related areas of secure communication, network building and hash functions. The lecture series will cover the necessary background on cryptography and point counting, and will introduce such tools as p-adic numbers, differential forms and Monsky-Washnitzer cohomology, from the ground up.
- ItemAn Assessment of select market timing strategies performance in Nairobi Securities Exchange(Strathmore University, 2019-08) Ng'ang'a, JudahHerding behavior among Kenyan traders in the capital markets has been majorly attributed to low levels of income and lack of knowledge in trading principles. This focus of this study was on the latter challenge. The study sought to avail information to traders in NSE regarding the profitability of select market timing strategies. The study evaluated the performance of three market timing strategies namely the relative strength index (RSI), simple moving averages (SMA) and hidden Markov model (HMM). The data considered in this study included the share price of the NSE-20 index over the period 2004-2018 triangulated to the perceptions and opinions of trading experts in Kenyan fund manager firms. The performance of market-timing strategies in this study was determined by a strategy's ability to generate above market returns as well as the accompanying Sharpe ratio. Using Henrikson and Merton market timing model, this study shows that all the market timing strategies have positive market timing strategies, with HMM having the best market timing ability. By accommodating the autoregressive nature of financial prices this study examined the ability of the strategies to time the market using an ARDL model. The model shows that all the strategies lack the ability to time the market but just like in HM model, HMM performances ranks best among the three strategies. Over the period 2004-2018 the Sharpe ratio of all the market timing strategies exceed that of SBH. The profitability is still the same when the period is split into two halves (2011-2018). The profitability persists when the trading points determined using the market index are used to trade five randomly selected securities. This confirms the profitability of trading based on patterns of historical prices. This negates the principles of EMH Theory but is in harmony with the FMH Theory, this study, therefore, provides support of FMH Theory as a more relevant market theory. In Kenya, this study found that investors are averse to market timing and tend to herd towards buy and hold strategy. Given the low profitability of the SBH strategy in NSE, this study should provoke the traders in NSE to reconsider their approach in determining optimal trading points.
- ItemBeliefs of secondary school mathematics teachers about problem solving skills and instruction practices(Strathmore University, 2019-08) Dorimana, Aline; Uworwabayeho, Alphonse; Nizeyimana, GabrielProblem solving Skill is for long highly considered as a skill that negotiates with the complex demand of 21st century labour market. Its importance is recognized by many countries around the world, including Rwanda. This is reflected in national education systems whereby national curriculums are consistently reviewed with regard to equip young children with problem solving skills at all level of education. Concerning with the case of Rwanda, educational curriculum for pre/primary and secondary levels are developed and reviewed at national level(Rwanda Education Board, 2015). At glance, it is recognized that teachers are the key component in allowing the learning to happen. The literature related to curriculum change points out that belief and perceptions of teachers seem to have considerable effects on the nature of classroom practice which is in line with student’s attitudes, beliefs and learning outcomes. Since the launch of the competence based curriculum in Rwanda, few studies if any have been conducted to understand how teachers perceive its requirements. The current study intends to provide insights about how Rwandan mathematics teachers view problem solving in the teaching and learning of mathematics after four years of the use of competence based curriculum that includes problem solving skill among generic skills that learners must acquire during their schooling. Specifically, the study will explore what upper secondary mathematics teachers believe and perceive about (a) Mathematical problem solving (b) Problem solving instructions (C) Problem solving in classroom practice (d) The kind of mathematics instruction emphasized in Rwandan Competence Based Curriculum (e) Difficulties about problem solving. In addition, the study will give insight on the relationship about beliefs and perceptions and classroom teaching practice. A questionnaire, interview and classroom observation will be used for data collection.
- ItemChebyshev-like polynomials satisfying fourth-order linear recurrences: Zeros and Hankel determinants(Strathmore University, 2019-08) Onyango, MichaelWe study sequences of polynomials that satisfy certain fourth-order linear recur rences with a parameter c. We show that for c real, their zeros lie on two concentric and inversely related circles. The associated n x n Hankel determinants are deter mined. Here, the 2 x 2 case is the most challenging, and has an intriguing connection with questions concerning sets of polynomials with all roots on the unit circle. These polynomials arise from Chebyshevian modifications of finite geometric series.
- ItemClassical vs Quantum: What does the future hold for fully homomorphic encryption schemes?(Strathmore University, 2019-08) Omollo, Richard; Odote, Jackson; Okello, BernardThe construction of homomorphic encryption schemes has supported arbirtrary computations on outsourced data, without revealing their true contents. Recent introduction of fully homomorphic versions has further improved the remedy to security challenges that affect data especially when not in custody of its owners, by supporting computations including those of higher polynomials functions. Despite these positive developments, implementation of these fully homomorphic encryption schemes has impacted negatively on the computing resources. These developments also have been made both for classical computing systems that are based on assumed hardness of mathematical problems and quantum computing systems that are based on Heisenberg principle. This paper aim at comparing the two sets of fully homomorphic encryption schemes: classical and quantum, with a view of providing future state-of-the-art of computational needs on management of consumer data. The output of this paper shall be useful not only to the scholarly world but also to the production units that shall desire to adopt a more efficient computational approach that inexpensively safeguard consumer data.
- ItemA Comparative analysis of rural and urban students’ performance in an Interactive Mathematics supported classroom(Strathmore University, 2019-08) Uwineza, Innocent; Uworwabayeho, AlphonsePerformance in education is a key characteristic of effective teaching and learning SBS and a criteria which is mostly used to classify schools from the best to the least. However, it depends on different factors including teaching resources and facilities. The current educational era is promoting the use of ICT in education as an instructional material with the purpose to raise the quality of education. After adopting the Competence Based Curriculum (CBC) which is now under its early implementation phase, the Rwandan education system is now seeking to find all the means to maximize the benefits of CBC to education. The development of the Interactive Mathematics (1M) content software for Rwanda is in this regard expected to bring the effectiveness of teaching and learning of mathematics in basic education by making mathematics enjoyable and accessible to children. After the testing of 1M in a bootcamp setting that had shown the 1M software potentiality to influence student’s motivation and performance, it had been necessary to analyse the situation in a real class setting. This paper provide the results of pre-test, post-test and comparison tests conducted in primary schools of rural and urban areas.
- ItemA comparative evaluation of goodness-of-fit tests for the negative binomial distribution with application to RNA-Seq data(Strathmore University, 2019-08) Osumba, John; Odhiambo, Collins; Omolo, BernardThe negative binomial (NB) distribution is considered the most appropriate distribution for modeling over dispersed count data. In this regard, a number of goodness of-fit (GOF) tests have been applied for the NB, but no systematic evaluation of these tests has been done to determine the most powerful tests. In this study, we perform a comparative evaluation of the GOF tests for the negative binomial distribution, based on their power under suitable alternatives, via simulations. The tests considered here include those based on the empirical distribution functions (EDFs), likelihood functions, Kullback-Leibler discrimination information, Laplace transforms, the generalized smooth tests and a combination of tests with complimentary behavior. For illustration and validation, RNA-Seq data from colorectal cancer are used.
- ItemCompetence Based Curriculum: Sticky issues for the uptake of STEM education and STEM careers in Africa(Strathmore University, 2019-08) Otieno, HerineIn the last decade, education policy in a many African countries has shifted to SBS Competence based education. Seen to resonate with and originate from global discourses on human capital theory and the current push to build a global workforce for the 21st century, the increased adoption of competence based education curriculum by African government continues to be fronted as key requirement for national economic growth. Using the proposed Kenyan Competence based curriculum as a case study this article provides a social-cultural theoretical reflection on some of the limitations of competency based curriculuxn as is currently promoted in some of the African countries in equitably supporting African students access to quality STEM education and transition STEM careers in the 21st century. Conclusions point to the need for bi/multilateral institutions such as the World Bank to desist from promoting one-sizefits all education models for Africa. A suggestion is made for key Pan-African STEM based institution such African Institute of Mathematical sciences (AIMS) to take a leadership role in advising the African governments on the merits and demerits of implementing a competency based curriculum.
- ItemConducting and communicating classroom-close research that makes an impact(Strathmore University, 2019) Dr Jennie, GoldingThese two sessions will complement Professor Jaworskis by focusing on two strands in research on mathematics teaching and learning: >As an insider in ones own institution, whether school, college or university >As an outsider and guest in other classrooms. At least the first is likely to be quite small-scale — but it can still have impact. Throughout, I hope we can all benefit by drawing on participants own experiences. In Session 1 we shall consider the rationale and audience for such research, its potential and limitations, and the implications of those for research planning and structure. I shall share my experiences and some of my learning from studies of both kinds, and participants will work in small groups to analyse the issues in relation to research they are doing, or would like to do. The particular methods adopted will depend on the context and research questions, but for all of them, ethics are key, so we shall work together to analyse the ethical and practical challenges associated with such work. In Session 2 we shall first consider the challenges associated with being an insider researcher. We shall then focus on the theoretical framings and tools that can be used to analyse and interpret classroom-close research, considering the ways those can be used to make meaning of a range of (often subjective) data and relate findings to existing literature. We shall also discuss means and benefits of developing interpretation collaboratively, including with participants. The second part of the session will focus on the purposes and form of dissemination: how can we communicate our findings to the range of audience we would want to know about them?
- ItemConservation laws and exact solutions of a generalized (3+1)-dimensional mKdV-ZK equation(Strathmore University, 2019) Adeyemo, Oke Davies; Masood, KhaliqueiIn this talk we investigate conservation laws and exact solutions of a generalized SBS (3+1)-dimensional mKdV-ZK equation that describes the behavior of weakly nonlinear ion-acoustic waves contained in magnetized electron-positron plasma which also comprises equal hot and cool components of each species. The use of the reductive perturbation procedure yields a mKdV-ZK equation that simply governs the oblique propagation of nonlinear electrostatic modes.
- ItemContent pricing in mobile social network: a collective bidding approach(Strathmore University, 2019-08) Awuor, MzeeAs smart mobile devices and phones become more ubiquitous and pervasive with wide array of sensors and communication techniques, we can develop mobile social network (MSN) apps that enable these devices to automatically create virtual communities where contents can be shared implicitly. For instance, your smartphone could assist you have a productive encounter with other MSN users by informing you about their interests and valuable contents that they may share with you. Example of such application is Whozthat which uses MSN to enrich offline social interactions among strangers by suggesting topics of common interest. Thus, MSNs enable users to discover and share contents with each other, especially at ephemeral events such as exhibitions and conferences. Nevertheless, the incentive of users to actively share their contents in MSNs may be lacking if the corresponding cost is high. In this paper, we propose a content pricing and sharing framework in MSN that is built on users’ collective bidding and content cost sharing. The content sharing problem is formulated as a distributed system that achieves cooperative outcome while preserving non-cooperative decision making among the users through the proposed collective bidding and broadcast nature of wireless communication. That is, co-located peers individually propose payments to their encounters whose contents they are interested in based on their perceived values of the contents. The respective content owners share their contents if the proposed payments can collectively compensate the cost of sharing their contents with these peers. We show that this guarantees individual rationality and promotes content sharing among the opportunistic encounters in the network. Performance evaluation shows that the proposed mechanism reduces the time and cost to collect contents of interest in the network and significantly improves network utilization.
- ItemCounting negative eigenvalues of one-dimensional Schroedinger operators with singular potentials(Strathmore University, 2019-08) Karuhanga, MartinIn this paper, we extend the well-known upper estimates of the Cwikel-Lieb-Rozenblum type for the number of negative eigenvalues of one-dimensional schroedinger operators with regular potentials to the case of strongly singular potentials. In particular, we consider the case when the potential is allowed to be a measure that is not necessarily absolutely continuous with respect to the Lebesgue measure.
- ItemCoverage probability of a non-parametric estimator for a finite population total using edgeworth expansion(Strathmore University, 2019) Okungu, Jacob; Orwa, George; Odhiambo, RomanusIn survey sampling, the main objective is more often than not to establish information about any population parameter using the sample statistics. A nonparametric estimator of the finite population total is proposed. The nonparametric estimator of finite population total by Dorfman (1992) is developed and the coverage probabilities explored using the Edgeworth. The asymptotic properties; unbiasedness, efficiency and coverage rates of the estimator are analytically explored. In literature, a lot of work has been done on analyzing unbiasedness and efficiency of the estimators and more particularly for the population total estimators. This study departs from these studies by studying the tail properties using the confidence interval in more detail as opposed to just the unbiasedness, efficiency and mean squared error. An empirical analysis is done on three artificial functions; linear, quadratic and exponentially. It is observed that the coverage probabilities from Edgeworth expansion have higher coverage probabilities compared to design-based Horvitz-Thompson and Ratio estimators of the finite population total. The Edgeworth expansion also gave a tighter confidence interval length.
- ItemCreating equitable environments through ethno-mathematics on a worldwide voyage(Strathmore University, 2019) Prof.Furuto, LindaEthno mathematics is a pathway to real-world problem solving that empowers locally minded, global citizens through a sense of purpose and a sense of place. Over the past decade, the Ethno mathematics and STEM Institute has explored learning and teaching practices via the design, implementation, and assessment of culturally sustaining work aligned with state and federal standards. Research has led to policy and practice implications including becoming institutionalized as a new academic program at the University of Hawaii. In this presentation I will highlight some of that research and also share the mission, vision and values that inspire the Ethno mathematics Graduate Certificate/M.Ed Curriculum Studies Math Education at University of Hawaii.
- ItemData driven longitudinal model with application to HIV differentiated care(Strathmore University, 2019-08) Odhiambo, Collins; Weunda, StephenDifferentiated care is a new innovative approach for managing HIV/AIDS where ART treatment services are customized by staggering patient’s visits for stable status while reducing unnecessary burdens on the health system. Through provision of differentiated care, the health system can reallocate resources to patients most in need who are failing treatment. The main objective of this study is to develop a data-driven longitudinal model which is applicable to HIV differentiated care. Method: We used routine data of HIV positive patients initiated to ART at the point of care from 4 medical facilities in Nairobi in the year 2018. Since both the GLMM and GEE are extensions of the GLM, we start with a brief overview of GEE then relooked at extensions of GLMM. We specify f (u) and g(p) to be dependent on the type of response Yi . For a binary Yi, we consider f (u) as Bernoulli distribution and g(p) as the logit function, g(u) — log resulting to GLM is the logistic regression. Results show the binary response which was differentiated care category fits well with GLMM. We also found TB-HIV co-infection to be the only significant predictor of differentiated care under both GEE and GLMM.