Clostridium difficile can cause severe diseases with significant morbidity and mortality in infected patients.
The rate of Clostridium difficile infection is high in North America and European countries. Metronidazole
and vancomycin have been recommended as the treatments of choice since 1990s. Recurrent infection
due to Clostridium difficile is common after several days of antibiotic administration. Probiotics have
been used in these patients as an adjunct treatment with some successful findings. However, a detailed
investigation on the use of probiotic for infected patients is still needed, particularly for its real efficacy.
As the amount of document increases, automation of classification that aids the analysis and management of documents receive focal attention. Classification, based on association rules that are generated from a collection of documents, is a recent data mining approach that integrates association rule mining and classification. The existing approaches produces either high accuracy with large number of rules or a small number of association rules that generate low accuracy. This work presents an association rule mining that employs a new item production algorithm that generates a small number of rules and produces an acceptable accuracy rate. The proposed method is evaluated on UCI datasets and measured based on prediction accuracy and the number of generated association rules. Comparison is later made against an existing classifier, Multi-class Classification based on Association Rule (MCAR). From the undertaken experiments, it is learned that the proposed method produces similar accuracy rate as MCAR but yet uses lesser number of rules.
Finding relevant disaster data from a huge metadata overhead often results in frustrating search
experiences caused by unclear access points, ambiguous search methods, unsuitable metadata, and long response times. More frequently, semantic relation between the retrieved objects is neglected. This paper presents a system architecture that makes use of ontologies in order to enable semantic metadata descriptions for gathering and integrating multi-format documents in the context of disaster management. After a brief discussion on the challenges of the integration process, the Multi-format Information Retrieval, Integration and Presentation (MIRIP) architecture is presented. A specific approach for ontology development and mapping process is introduced in order to semantically associate user’s query and documents metadata. An ontology model approach was designed to follow inspirational and collaborative approaches with top-down to bottom-up implementation. A prototype of the integrated disaster management information system is currently under development, based on the architecture that is presented in this paper.
Bananas were chilled at 6oC and the appearance of brown spots when exposed to ambient air, a
phenomenon known as chilling injury (CI), was detected using computer vision. The system consisted of a digital colour camera for acquiring images, an illumination set-up for uniform lighting, a computer for receiving, storing and displaying of images and software for analyzing the images. The RGB colour space values of the images were transformed into that of HSI colour space which is intuitive to human vision. Visual assessment of CI by means of a browning scale was used as a reference and correlation between this reference values and hue was investigated. Results of the computer vision study successfully demonstrate the potential of the system in substituting visual assessment in the evaluation of CI in bananas. The results indicate significant influence, at α=0.05, of treatment days and temperature on hue. A strong correlation was also found between hue and visual assessment with R>0.85.
This paper is concerned with the existence, types and the cardinality of the integral solutions for
diophantine equation
4 4 3
x y z + = where x , y and z are integers. The aim of this paper was to
develop methods to be used in finding all solutions to this equation. Results of the study show the
existence of infinitely many solutions to this type of diophantine equation in the ring of integers
for both cases, x y = and x y ¹ . For the case when x y = , the form of solutions is given by
3 3 4
( , , ) (4 , 4 ,8 ) x y z n n n = , while for the case when x y ¹ , the form of solutions is given by
3 1 3 1 4 1
( , , ) ( , , )
k k k
x y z un vn n
- - -
= . The main result obtained is a formulation of a generalized method to find
all the solutions for both types of diophantine equations.
Heuristic Evaluation (HE) is used as a basis in developing a new technique to evaluate usability or
educational computer games known as Playability Heuristic Evaluation for Educational Computer Game (PHEG). PHEG was developed to identify usability problems that accommodate five heuristics, namely, interface, educational elements, content, playability and multimedia. In HE process, usability problems are rated based on severity score and this is followed by presentation of a mean value. The mean value is used to determine the level of usability problems; however, in some cases, this value may not accurate because it will ignore the most critical problems found in a specific part. In developing PHEG, a new quantitative approach was proposed in analyzing usability problems data. Numbers of sub-heuristics for each heuristic involved were taken into account in calculating percentage for each heuristic. Functions to calculate critical problems were also introduced. Evaluation for one educational game that was still in development process was conducted and the results showed that most of the critical problems were found in educational elements and content heuristics (57.14%), while the least usability problems were found in playability heuristic. In particular, the mean value in this analysis can be used as an indicator in identifying critical problems for educational computer games.
Similarity measurement is a critical component in any case-based reasoning (CBR) system. CBR is
a superior technique for solving new problems based on previous experiences. Main assumption in
CBR relies on the hypothesis that states similar problems should have similar solutions. This paper
describes a comparative analysis on several commonly used similarity measures (Canberra, Clark, and Normalized Euclidean distance) in retrieving phase of the case-based reasoning approach to facilitate supplier selection. In addition, the proposed agent-based supplier selection framework was designed to use customer’s defined weights to evaluate the price, volume, quality grade, and delivery date of supply materials, and also provide them with alternative products which are closest to their first order if it was out of stock. Finally, based on the proposed framework, a numerical example of the used approach is illustrated.
Mobile users have the capability of accessing information anywhere at any time with the introduction of mobile browsers and mobile web search. However, the current mobile browsers are implemented without considering the characteristics of mobile searches. As a result, mobile users need to devote time and effort in order to retrieve relevant information from the web in mobile devices. On the other hand, mobile users often request information related to their surroundings, which is also known as context. This recognizes the importance of including context in information retrieval. Besides, the availability of the embedded sensors in mobile devices has supported the recognition of context. In this study, the context acquisition and utilization for mobile information retrieval are proposed. The “just-in-time” approach is exploited in which the information that is relevant to a user is retrieved without the user requesting it. This will reduce the mobile user’s effort, time and interaction when retrieving information in mobile devices. In this paper, the context dimensions and context model are presented. Simple experiments are shown where user context is predicted using the context model.
Cross-document Structure Theory (CST) has recently been proposed to facilitate tasks related to multidocument analysis. Classifying and identifying the CST relationships between sentences across topically related documents have since been proven as necessary. However, there have not been sufficient studies presented in literature to automatically identify these CST relationships. In this study, a supervised machine learning technique, i.e. Support Vector Machines (SVMs), was applied to identify four types of CST relationships, namely “Identity”, “Overlap”, “Subsumption”, and “Description” on the datasets obtained from CSTBank corpus. The performance of the SVMs classification was measured using Precision, Recall and F-measure. In addition, the results obtained using SVMs were also compared with those from the previous literature using boosting classification algorithm. It was found that SVMs yielded better results in classifying the four CST relationships.
Developing a trust management scheme in mobile computing environment is increasingly important,
and the effective trust management model is a challenging task. Business, education, military, and
entertainment have motivated the growth of ubiquitous and pervasive computing environments, which are always available due to the widespread of portable and embedded devices. Wireless and mobile computing are good example of ubiquitous and pervasive computing environments. Due to the uncertainty and mobility in such environments, the issue of trust has been regarded as an important security problem. Malicious nodes are a major threat to these networks; the trust system can monitor the behaviour of nodes and accordingly rewards well-behaved nodes and punishes misbehaving ones. At present, there are a lot of endeavours on the trust model of the pervasive computing environment. In this paper, a trust management framework for mobile computing is presented. The hybrid framework is based on a fusion of the support vector machine (SVM) and fuzzy logic system. From the results, it can be stated that the framework is effective, dynamic, lightweight, and applicable.
Natural language interfaces to ontologies allow users to query the system using natural language queries.
These systems take natural language query as input and transform it to formal query language equivalent
to retrieve the desired information from ontologies. The existing natural language interfaces to ontologies
offer support for handling negation queries; however, they offer limited support for dealing with them.
This paper proposes a negation query handling engine which can handle relatively complex natural
language queries than the existing systems. The proposed engine effectively understands the intent of
the user query on the basis of a sophisticated algorithm, which is governed by a set of techniques and
transformation rules. The proposed engine was evaluated using the Mooney data set and AquaLog dataset,
and it manifested encouraging results.
The 2004 Banda Aceh earthquake and ensuing Andaman mega tsunami that killed a quarter million people worldwide is a wake-up call to many. Active research was initiated in Universiti Sains Malaysia (USM) immediately after the infamous event with the aims to help develop human capacity and resources, and to mitigate any future earthquake and tsunami. The Disaster Research Nexus (DRN) was formed recently within the School of Civil Engineering, USM, to facilitate active collaborative research on earthquakes and tsunamis, as well as on other natural disasters, such as landslides. This paper begins with an introduction to DRN. This is followed by a description of some research achievements undertaken by DRN staff. A concise exposition on the tsunami simulation model TUNA developed by the authors and its application to the 2004 Andaman tsunami are given to illustrate the capability of TUNA. The role of mangrove in reducing the impact of tsunami is then modelled. Tsunami may inundate coastal plain with large quantity of saline water, changing the salinity regimes in the soil and inducing vegetative succession changes. A model called MANHAM was developed to simulate the salinity changes and its associated vegetative evolution to assist in the rehabilitation of vegetation destroyed by tsunami. Meanwhile, an earthquake risk analysis for the Upper Pandas Dam in Sabah is then presented, and this is followed by a model estimation of tsunami forces on the coastal structures. The main objective of this paper is to reach out to research scientists and onsite risk reduction professionals to collaborate towards the development of a vibrant research culture to face future natural disasters such as earthquakes and tsunamis. It is hoped that DRN will move forward to further enhance active collaborations with other research and operational institutions worldwide towards developing earthquake and tsunami resilient communities.
Research on the use of Jatropha curcas triglycerides as biodiesel feedstock has received worldwide attention due to its inherent characteristics. Unlike palm oil, J. curcas oil is not edible, and thus, it will not disturb the food supply. However, to the researchers' experiences with the synthesis of J. curcas, oil-based biodiesel has shown that the fuel characteristics depend largely on the type of alcohol used as the excess reactants. Transesterification reaction is chosen for this process with sodium methoxide as the catalyst. Comparison studies on the yield of esters using methanol and ethanol, as well as the impacts on the reaction rate are discussed. The effects of reaction time and molar ratio on the reaction conversion are also examined. The determination of reaction yield is based on the conversion of triglycerides into alkyl esters as the main product. The findings are described as follows: the highest percentage yield of product is attained at 96% for methanol as an excess reactant, and this is 90% when ethanol is used. The optimum conditions of parameters are achieved at 6:1 molar ratio of alcohol to triglycerides, 50 min of reaction time and reaction temperature of 65°C for methanol and 75°C for ethanol. The biodiesel properties of both ester fuels were determined according to the existing standards for biodiesel and compared to the characteristics of diesel fuel.
It is now evident that the estimation of logistic regression parameters, using Maximum LikelihoodEstimator (MLE), suffers a huge drawback in the presence of outliers. An alternative approach is touse robust logistic regression estimators, such as Mallows type leverage dependent weights estimator(MALLOWS), Conditionally Unbiased Bounded Influence Function estimator (CUBIF), Bianco andYohai estimator (BY), and Weighted Bianco and Yohai estimator (WBY). This paper investigates therobustness of the preceding robust estimators by using real data sets and Monte Carlo simulations. Theresults indicate that the MLE behaves poorly in the presence of outliers. On the other hand, the WBYestimator is more efficient than the other existing robust estimators. Thus, it is suggested that the WBYestimator be employed when outliers are present in the data to obtain a reliable estimate.
The incidence of ureteric calculus as a cause for severe abdominal pain in children is mounting, especiallyin the tropical country. The course of illness may be non-specific but a swift detection via non-invasiveimaging modalities singly or in combination may avert unnecessary radiation hazard and futile surgery ina young child. In this paper, we discussed a case of an 11 year-old boy who was presented with a suddenonset of the right side severe abdominal colic whose a bedside ultrasound was positive for hydronephrosisfor which localisation of stone was further confirmed via a low dose limited intravenous urography (IVU).It is important to note that data available on the value of a combined ultrasound and the limited IVU inan emergency setting when urolithiais is being suspected in children with abdominal pain are particularlyscarce. Hence, this case documented the potential value of a combined ultrasound and a limited IVUstudy as a unique combined armamentarium used in a suspected childhood urolithiasis in the tropics.
Kenaf fibre that is known as Hibiscus cannabinus, L. family Malvaceae is an herbaceous plant that can be grown under a wide range of weather conditions. The uses of kenaf fibres as a reinforcement material in the polymeric matrix have been widely investigated. It is known that epoxy has a disadvantage of brittleness and exhibits low toughness. In this research, liquid epoxidized natural rubber (LENR) was introduced to the epoxy to increase its toughness. Kenaf fibres, with five different fibre loadings of 5%, 10%, 15%, 20% and 25% by weight, were used to reinforce the epoxy resins (with and without addition of epoxidized natural rubber) as the matrices. The flexural strength, flexural modulus and fracture toughness of the rubber toughened epoxy reinforced kenaf fibre composites were investigated. The results showed that the addition of liquid epoxidized natural rubber (LENR) had improved the flexural modulus, flexural strength and fracture toughness by 48%, 30%, and 1.15% respectively at 20% fibre loading. The fractured surfaces of these composites were investigated by using scanning electron microscopic (SEM) technique to determine the interfacial bonding between the matrix and the fibre reinforcement.
Oil palm is widely grown in Malaysia. There has been interest in the utilization of oil palm biomass for production of environmental friendly biofuels. The gasification of empty fruit bunches (EFB), a waste of the palm oil industry, was investigated in this study to effectively and economically convert low value and highly distribution solid biomass to a uniform gaseous mixture mainly hydrogen (H2). The effects of temperature, equivalence ratio (ER) and catalyst adding on the yields and distribution of hydrogen rich gas products were also investigated. The main gas species generated, as identified by GC, were H2, CO, CO2, CH4 and trace amounts of C2H4 and C2H6. With temperature increasing from 700 to 1000 °C, the total gas yield was enhanced greatly and reached the maximum value (~ 90 wt. % ) at 1000°C with a big portion of H2 (38.02 vol. %) and CO (36.36 vol. %). Equivalence ratio (ER) showed a significant influence on the upgrading of hydrogen production and product distribution. The optimum ER (0.25) was found to attain a higher H2 yield (27.42 vol. %) at 850°C. The effect of adding catalysts (Malaysian dolomite1, P1), Malaysian dolomite2 (GML), NaOH, NaCl, CaO, ZnO, NiO) as a primary catalyst on gas product yield was investigated, and it was found that adding dolomite showed the greatest effect with the maximum H2 yield achieved (28.18 vol.%) at 850°C.
As a crucial demand in urban areas, flood risk management has been considered by researchers and decision makers around the world. In this case, hydrological modelling that simulates rainfall-runoff process plays a significant role. This paper quantified the roles of three main parameters in river basin hydrological response, namely, rainfall event duration, rainfall event ARI (magnitude) and land-use development condition. The case study area of this research was Sungai Kayu Ara basin which is located in the western part of Kuala Lumpur, Malaysia. A total of twenty seven scenario were defined for this research, including three different rainfall event durations (60, 120 and 360 minutes), three different ARIs (20, 50 and 100 years) and in three different land-use conditions (existing, intermediate and ultimate). The results of this research indicate that rainfall event duration, rainfall event ARI (magnitude) and land-use development condition have considerable effects of the surface runoff hydrographs in terms of peak discharge and volume.
In the literature, several methods of ground improvement have been presented including compacted stone columns. The bearing capacity of the granular column is governed mainly by the lateral confining pressure mobilized in the soft soil to restrain or prevent bulging of the granular column. Therefore, the technique becomes unfeasible in peat that does not provide sufficient lateral confinement. This condition can be overcome by encasing the stone column with geogrid. This paper investigates the performance of the geogrid encased vibrocompacted stone column in peat. This study was carried out using PLAXIS software equipped with unit cell concept. The peat was modelled using soft soil model and the stone column using Mohr-Coulomb soil model, respectively. The geogrid was modelled using the geogrid option and could take only tensile force. The results indicate that the geogrid encased stone column can take much higher load in comparison to ordinary stone columns as the stiffness of the column increases. Meanwhile, the length of encasement also varied and it was observed that it was very effective up to about two times the diameter of the column. It also increased the column stiffness, and therefore led to a significant strain reduction. It was also observed that the columns at a spacing of three times the diameter are very effective. The results presented here can be used by the geotechnical engineers to design the geogrid reinforced stone column based on the strength of the soil, diameter of the column, spacing of the columns and stiffness of the geogrid.
Wind energy has often been touted as one of the most reliable sources of renewable energy that should be used for people. Today, wind energy (mainly by propeller type wind turbines) produces less than one percent of the total energy used worldwide. Practically, a standard three-blade propellers efficiency of use of the wind energy is around twenty percents and this is due to its design and shape that use the wind lift force and a rotating turbine. In addition, these turbines are quite expensive due to the complex aerodynamic shape of the propellers which are made of composite materials. The new world boom for wind turbines obliges inventors to create new wind turbine designs that have high efficiency and are better than any known design. This paper proposes the new patented invention of the vane-type wind turbine which uses wind energy more efficiently and is only dependent on the acting area of the vanes. The vane wind turbine was designed to increase the output of a wind turbine that uses kinetic energy of the wind. Due to its high efficiency, simple construction and technology, the vane wind turbine can be used universally, apart from the fact that it is made from cheap materials. The new design of the vane-type wind turbine has quite small sizes than the propeller type one of same output power.