This study investigates the use of Facebook as a Learning Management System on junior and senior students doing multimedia-based courses. The study uses a conceptual model based on the second thesis of Andersons Interactivity Equivalency Equation to ascertain the impact the use of the technology has on student learning outcomes. Results showed that the most significant factor of interaction was the student to teacher construct. Due to their interaction on the social media site group with their peers, content and teacher, they achieved higher final results. Significant factors were the value placed on real-time chatting with peers, practical assessment on overall learning and the reading of instructor posts. 75% of the control group had GPA less than 2.70, which suggest that the lack of the additional interaction may have been a contributing factor to lower GPAs than the experimental group which had a mean GPA of 3.71.
In current era of Nanotechnology Through silicon vias (TSVs) have potentially provided an attractive solution for the development of reliable 3D integrated system. The 3D integrated system is potentially dependent on the filler materials used in TSVs. This research paper introduces multi walled carbon nanotubes (MWCNT) bundle and Multillayered graphene nanoribbons (MLGNRs) as filler materials for TSVs. Equivalent single conductor (ESC) model for different bundle configuration is employed to analyze the propagation delays. It is observed that at lower technology node, the overall delays are reduced by 6.311% and 10.86% respectively for MWCT bundle and MLGNR compared to higher technology node.
Various classification and supervised learning techniques can be applied to nonlinear practical problems. Electroencephalographic (EEG) frequency classification using discriminant kernels was used as an elegant classification technique in analysing the forward inverse problem in EEG artefact selection. Discriminant kernels masked in Kaczmarz and Algebraic Reconstruction Technique used EEG frequency feature to learn and transform the EEG signal during EEG artefact selection. The effective management of EEG data through linearization of the artefact selection model provided an efficient local optimal solution in brain-computer interface development and robot control signal application. The models were used as a microcontroller resource management tool in EEG data management and robot control signal selection. This paper presents the selection of EEG artefact using linear discrimination kernels in analysing the EEG inverse problem. The paper also discussed the implementation of Kaczmarz's and algebraic reconstruction technique (ART) in EEG data management. The selected EEG artefacts were used as robot control signals.
The city of Johannesburg has recently constructed and demarcated cycling lines alone some of its major highways. Some of them connect the University of Johannesburg campuses which are across towns. None of the University staff and students seems to have any interest in cycling. The study is based on quantitative research. Questionnaires were given to students and 484 students responded. The study aimed to investigate if the student are aware of the cycling lanes. Through the response from first survey question, students seem to be concern about safety. Students were also asked if in their opinion cycling is safe. 50% of the response said no while the other half said yes. Out of the number that said yes there where asked how is cycling unsafe. Students also gave their opinions and also choose from theft (in their residence), Theft elsewhere (e.g., school, shopping Centre), reckless driving, cycling lane design and surface, lack of cycling skills and mugging and robbing. 484 students responded to this study. Reckless driving is ranking at 86% percent which shows that the respondents are more concerned of the motorist who ignores or do not respect the rules of the road.
Image Segmentation is a process of partitioning a digital image into multiple segments. Retinal Vessel segmentation is a procedure to extract the various blood vessels to diagnose various diseases such as diabetic retinopathy. Most of the diabetic patients generally have retinopathy disorders that can be seen by vessel segmentation. In this paper, various techniques have been evaluated and it has been observed that retinal vessel segmentation using hybrid filters is the best technique in terms of accuracy.
Conventional Travelling Salesman Problem (TSP) solutions are based on the assumption that time of travel between nodes is wholly dependent on distance. But in practice, this is not so as road and traffic conditions help to determine time taken to travel between nodes. We introduce fuzzy based TSP solution where distance and road traffic conditions are fuzzy. Result of the fuzzy based solution for a case study show that using conventional TSP solution the average time to travel shortest path was 192min while with the fuzzy TSP solution, the average time was 177.66min. The result confirms the superiority of fuzzy TSP solutions over conventional TSP in solving real life TSP problems.
The City of Johannesburg, a world class African city is looking to improve air quality, cut down on greenhouse gas emissions and generate alternative sources of fuel by creating an effective management system for the enormous amount of waste generated within the city. The structures put in place would not only cut down dependence on landfills but provide fuel to power the city's metro buses and in the nearest future generate electricity. This paper focuses on the quantification exercise at the Johannesburg Market which has the potential to generate about 28, 486 GJ of energy per year from the anaerobic digestion of fruit and vegetable waste which accounts for 93% of the total waste generated at the market.
Electric discharge machining (EDM) process generally used for burrs free, less metallurgical damage, stress free and very precise machining and produces mould cavity, deep holes, complex shapes & size by arc erosion in all types of electroconductive materials. In this process, the metal is removed from the work piece due to erosion caused by rapidly recurring spark discharge taking place between the tool electrode and work-piece. Tool electrode and wok-piece both submersed into the dielectric fluid. The main aims of this review paper work is to present the consolidated information about the contribution of various researchers on the machining applications of electric discharge machining process on Nickel-Base Super alloys materials, utilization of various tool and techniques for correlating experiment results and applications of product through the EDM. Nickel-Base Super alloys materials is widely used for fuel tanks, aircraft & rocket engine components, nuclear fuel element spacers, casings, fasteners, rings, seal, measuring instrument, cryogenic storage tanks and automobile components etc.
This paper presents a complex approach for modelling of a tandem reservoir systems for water drainage management. The model has been built over a segment of a river with a certain parameters of water inflow, water outflow, required power production and max possible flood occurrence. Then the segments may be replicated with specific parameters to simulate whole system of the river. The model has been optimized in order to obtain the water drainage operation policy with regards of current and expected water volumes in the reservoirs ratio, required power production revenue, and minimal flood occurrence. Model has been verified on a walk-through basis. The obtained results demonstrate good reliability disregards broad possible variations of the managed parameters and provide the optimal water drainage for minimum flood occurrence and desired power production revenue.
Let m, r, k be three positive integers. Let G be a graph with vertex set V(G) and edge set E(G), and let f: V(G) → N be a function such that f(x)≥(k+2)r−1 for any x ∈ V(G). Let H1, H2, …, Hk be k vertex disjoint mr-subgraphs of a graph G. In this paper, we prove that every (0,mf−(m−1)r)-graph admits a (0, f)-factorization randomly r-orthogonal to each Hi (i=1,2,…,k).
Hydrogen is expected to be one of the most important fuels in the near future for solving the problem caused by the greenhouse gases, for protecting environment and saving conventional fuels. In this study, a dual fuel engine of hydrogen and diesel was investigated. Hydrogen was conceded through the intake port, and simultaneously air and diesel was pervaded into the cylinder. Using electronic gas injector and electronic control unit, the injection timing and duration varied. In this investigation, a single cylinder, KIRLOSKAR AV1, DI Diesel engine was used. Hydrogen injection timing was fixed at TDC and injection duration was timed for 30°, 60°, and 90° crank angles. The injection timing of diesel was fixed at 23° BTDC. When hydrogen is mixed with inlet air, emanation of HC, CO and CO2 decreased without any emission (exhaustion) of smoke while increasing the brake thermal efficiency.
The operational aircraft maintenance routing problem with flight delay consideration (OAMRPFD) determines the route to be flown by each aircraft in real aspect life. It is observed that OAMRPFD related studies was formulated based on the expected value of the non-propagated delay, which is any delay caused by non-routing issues such as bad weather, technical problems, passengers delays, etc. However, a drawback of this formulation is that the expected value approach may not adequately reflect the final realization of non-propagated delay, as it is characterized by high level of uncertainty. This would result in facilitating the propagation of the delay and increasing its related cost paid by the airline companies. In this paper, we study OAMRPFD with an objective of developing an OAMRPFD that reflects appropriately the final realization of non-propagated delay. For this purpose, a new scenario-based stochastic framework for OAMRPFD (SOAMRPFD) is proposed. In order to solve the proposed model, an Ant Colony Optimization (ACO) algorithm is proposed. A case study of major airline company located in the Middle East is presented to demonstrate the potential of the proposed model.
In engineering project management, it is necessary to form teams with persons to accomplish plans. For many teams leaders, the team members selection could be a real challenge, due to the complex problem of set up a productive unit. An essential requirement is teamwork skill, especially in engineering workgroups where the project member is expected to know how to collaborate with peers. In this work-in-progress paper, we used social networks to represent social links between team members prospects. In our case of study, undergraduate computer engineering courses, the students expressed their preferences for working with other three peers at the course beginning, and this information was used to distinguish groups within the social network using network analysis algorithms. We compared the network analysis results versus groups formed by a teacher in a real course. Finally, we discussed the advantages and disadvantages of project teams from social network analysis approach to making team formation recommendations into a socio-technical system.
Montmorillonite K10 (MMT K10), the major clay mineral commercially available can be treated to improve its properties and thus can be utilized for a broad range of organic reaction including esterification of fatty acids for biodiesel production. However, the use of unmodified MMT K10 in esterification of fatty acids showed that the acid conversion was less than 60 %. The aim of this study is to utilize available material i.e. MMT K10 modified with Cu2+, Al3+ and Fe3+ at various concentrations as catalyst for the esterification of acetic acid and stearic acid. The x-ray diffraction and elemental analysis of the materials were successfully characterized. After characterization, the materials were evaluated for the esterification of acetic acid and stearic acid with methanol. Prepared catalysts were able to give the highest acid conversion of up to 75% relative to the unmodified MMT K10 revealing the potential of M-MMT K10 as catalyst for esterification in biodiesel production.
In this study, grinding operation was performed on a work-piece with an unknown shape. As the control strategy, hybrid force/velocity control method was utilized using a PID controller and an Active Disturbance Rejection Controller (ADRC). Both control structures were implemented on an experimental setup and the results were compared. Results show that significant amount of effort can be saved in grinding operation on a work-piece with an unknown shape due to the elimination of CAD model dependent path planning processes.
Soft computing deals with imprecision, uncertainty, partial truth, and approximation to achieve practicability, robustness and low solution cost. As such it forms the basis of a considerable amount of machine learning techniques such as Artificial Neural Network, Support Vector Machines, Fuzzy Logic, Evolutionary algorithm and Swarm Intelligence. Oil and gas investment project is comprehensive and capital intensive engineering system characterized by uncertainty and high risk. The decision to invest is usually based on the evaluation of the project profitability which in most cases is not accomplished with high level of certainty. These economic evaluations are carried out with estimated set of parameters such as project costs, oil and gas prices, production profiles and inflation rate. The paper considers an approach to apply the concept of soft computing to uncertainty analysis in the economic analysis of oil and gas investment. A description of methods and models used for economic evaluation are provided. The application of soft computing such as support Vector Machines and Fuzzy Logic models are also provided. The soft computing approach is compared with basic evaluation without consideration for uncertainty. A case study is presented to illustrate that the data from which the evaluation is drawn are limited leading to an estimate that are inherently uncertain. The motivation of the proposed method is the fact that that the hybrid Support Vector Machines and Fuzzy Logic models have been used to predict oil prices, cost, net present value and internal rate of return with reasonable level of certainty. It also enables adequate risk assessment inherent in an investment to facilitate appropriate decision making.
Fischer-Tropsch synthesis (FTS) was explored by combining a non-thermal plasma (NTP) with a 0 (blank), 2 or 6 wt%-Co/A1203 mullite catalyst at very high pressure (0.5 to 10 MPa) and at different treatment periods of 10 and 60 s. The 6 wt% Co catalyst system produced the highest methane, ethane, ethylene and propane yields at 2 MPa and 60 s, which were similar to the yields for the 2 wt% Co catalyst and 46, 96, 270 and 25 times higher than that of pure plasma.
The issue of automatic learning of the morphology of natural language is an important topic in computational linguistics. This owes to the fact that morphology is foundational to the study of linguistics. In addition, the emerging information society demands the application of Information and Communication Technologies (ICT) to languages in ways that demand human-like analysis of language and this depends to a large extent on the ability to undertake computational analysis of morphology. Even though rule-based and supervised learning approaches to the modeling of morphology have been found to be productive, they have also been discovered to be costly, cumbersome and sucseptible to human errors. Contrarily, unsupervised learning methods do not require the expensive human intervention but as in everything statistical, they demand large volumes of linguistic data. This poses a challenge to resource scarce languages such as Igbo. Furthermore, being a highly agglutinative language, Igbo features certain morphological processes that may not be easily accommodated by most of the frequency-driven unsupervised learning models available. this paper takes a critical look at some of the identified challenges of inducing Igbo morphology as a first step in devising methods by which they can be addressed.
Existing research focus on extracting the concepts and relations within a single sentence or in subject-object object pattern. However, a problem arises when either the object or subject of a sentence is "missing" or "uncertain", which will cause the domain texts to be improperly presented as the relationship between concepts is no extracted. This paper proposes a solution for the enrichment of the knowledge of domain text by finding all possible relations. The proposed method suggests the appropriate or the most likely term for an uncertain subject or object of a sentence using the probability theory. In addition, the method can extract the relations between concepts (i.e. subject and object) that appear not only in a single sentence, but also in different sentences by using a synonym of the predicates. The proposed method has been tested and evaluated with a collection of domain texts that describe tourism. Precision, recall, and f-score metrics have been used to evaluate the results of the experiments.
The main purpose of this paper is to extend the reasons why linear equation systems, under the appropriate conditions, must be solved by their graph-based representation. Such representation in the solution of systems of linear equations is an efficient alternative to the matrix representation. Among its benefits we have the sparse management of the information of the system in question in a natural way, and as will be explained in this document, it gives rise to active and non-active regions that if they are attacked in the traditional way through i.e. matrix, the algorithms developed will be inefficient. This type of systems appears naturally in problems of nonlinear programming and therefore its knowledge and especially application can make the solution of a Nonlinear Programming problem more efficient.
