The government, non-government organizations and community leaders in some countries were challenged to design and implement policies, programs and systems that could help local communities to overcome various threats. Changes of situation in the community, such as relocation, was considered a threat. Relocation has been regarded an unpopular plan as it raised controversial issues among the affected communities, leading to conflicts. In Indonesia, especially DKI Jakarta, in order to implement the Urban Environment Management Program, the Government of DKI Jakarta conducted land control to refunction some land into green open space, in order to bring the function of the Ciliwung River back to normal, and to control the land under the possession of PT. Indonesian Railways and as to revitalize the marine tourism areas. The government relocated the community who occupied illegal land to simple rental flat (rusunawa). People who were impacted by this policy had to adjust themselves to the new environment to make living, and some other chose to stay in a modest flat. The concept of community resilience has been an interesting issue as it dealt with the ability of something, someone or even society in overcoming difficulties, to recover and return to normal conditions at the right time. This study was done to propose a conceptual framework for community resilience for the residents of modest rental flats and to determine a standard method to examine the community resilience among residents of modest rental flat who were impacted by the relocation policy. Literature review was done to obtain secondary data and relevant scientific literature. In addition, observations were done to several modest flat which were established prior to the relocation policy (Marunda, Rawa Bebek and Kapuk Muara Penjaringan). To measure the strength and direction of the independent variable's contribution to the dependent variable, Variance Based Structural Equation Modeling was administered using the Smart-PLS program. The proposed conceptual framework for community resilience among the residents of modest flat covered 8 (eight) aspects, namely: 1) economy; 2) social; 3) culture; 4) human resources; 5) ecology; 6) physical; 7) politics and government and 8) technology. Those aspects should be taken into consideration in order to increase the capacity of modest apartment to stay. For the government or policy makers and information planners, the proposed framework can be used as an input for policy making, determining the priority scale, fund allocation and the enhancement of the sustainable flat housing development programs.

The role of non-conventional yeast Kluyveromyces marxianus in bioprocessing industry has shown potential as metabolites producer, make it a suitable candidate in replacing baker's yeast for various industrial applications. Mathematical approach is used to analyze the flow of metabolites in biological system in order to improve desired product yield as well as the overall production process, thus development of a simple model could lead to sustainability and practicability of the process. In this study, the comparative analysis of a simple metabolic network and a black box description is carried out in order to evaluate the yeast growth and bioethanol production in K. marxianus batch culture. Metabolic flux balance methodology has shown to give more accurate estimation with complete analysis of the reaction rates. Furthermore, better evaluation on batch yeast behavior and performance at varied glucose concentrations was achieved based on its stoichiometric reaction analysis. The highest biomass growth (12.32 g/l) was found compared to bioethanol production as the highest substrate concentration supplied in the medium. The biomass and bioethanol production were mostly relied on oxidative and reductive catabolism, respectively, in which the glucose and oxygen uptake rates play the main role in regulation of the central metabolic networks. Therefore, either of these products is strongly reliant on the cellular functionality of yeast in the culture, which shows the superiority of this method over black box approach.

The potential utility of epoxidized vegetables oil has begun to be realized in industrial application with increasing the concern of research to develop value added products from the available vegetables oil. Epoxidized oleic acid is often regarded as a high value oleochemical due to wide range of industrial applications including cosmetic, lubricants, painting and as chemical precursor. The aim of this study is to optimize the epoxidation of oleic acid of palm oil. The method was evaluated on different parameters such as oxygen carrier, concentration of catalyst as well as different type of vegetables oil. The optimum relative conversion to oxirane (86%) was achieved for 15 minutes of reaction time. Epoxidation results were based on complete conversion oxirane, rate of epoxidation and stability of the oxirane.

The epoxidized of vegetable oil is one of the important chemical processes to produce the commercial goods such as plastics, lubricants, and paints. A good kinetic model is necessary to predict the process outcome, especially for the big scale productions in which very helpful in predicting the actual production. In this study, the kinetic model of the epoxidation process of palm kernel oil was investigated. The kinetic model was developed based on the laboratory works done by Jumain [1]. The mathematical model was established and implements into the MATLAB by integrating numerically using fourth order Runge-Kutta method. The reaction rates parameters denote as k11, k12, k13, and k22 were then estimated using the Hooke-Jeeves algorithm until the convergence values obtained. After 100 increments, the reaction rates parameters are k11 = 0.1875 mol/L.min, k12 = 0.9997 mol/L.min, k21 = 0.0625 mol/L.min and k22 = 0.0000 mol/L with an absolute error of 0.0857. Good agreement was found between experiment and simulation. Based on the converged kinetic values, the evolution of the con-centration for all the species was simulated. After the end of the simulations, the computed concentration of the epoxide is approximately 1.5169 mol/L.

Gross Domestic Product (GDP) has been known as multi criteria measurement of market value from goods and services across nation. Even there is a improvement study on accuracy recently, this subject still opens a huge potential of future researches and method develop-ment and implementation. At this study we compared a new approach on Moving Average method called Weighted Exponentially Moving Average (WEMA) that created in 2014. We implements the method on Phatsa Web Application and evaluate the results by calculating its Mean Squared Error (MSE), Mean Absolute Scaled Error (MASE), and Mean Absolute Percentage Error (MAPE). Comparison study with conventional moving average named Simple Moving Average (SMA), Weighted Moving Average (WMA) and Exponentially Moving Average then conducted to review the capability of WEMA in predicting Indonesia's GDP data. It is proven that WEMA can provide more accurate result rather than SMA and WMA, but it still has the same error rate value if compared to EMA with more computation time. In addition to the result, Czech's GDP dataset is also tested and provides more information on how WEMA calculation with expanded dataset and more volatile value compared to Indonesia's GDP in Czech's GDP dataset.

The development of Web 2.0 technology affects the communication and collaboration processes carried out by its users. The use of social technology in the formal organizational environment and for formal needs is influenced by several factors. The TOE and STS aspects are used to understand the use of this technology for formal learning purposes that are carried out independently or self-learning. The TOE aspect is used to analyze technology components in relation to the organization and environment that uses it. The STS aspect is used to see the suitability of the technology to the task and its users including from the social side. Acceptance of social technology attached to web 2.0 technology by students by utilizing for the learning process. The use of social media for the self-learning process shows that perceived behavior control does not significantly affect the intention of its users. User intention on social media in the self-learning process is influenced by the attitude and subjective norms of its users. The social influence brought by web 2.0 technology influences the subjective norms of its users.

The advantage competitive is always being issue to improve the capability of the institution even the higher education. So the higher educa-tion needs to evaluate its capabilities in facing this issue so the higher education could determine its strategic plan. This paper aims at apply-ing the Tsukamoto method in fuzzy logic to measure the implementation of strategic plan that combined with the valuable, rareness, inimita-bility, and Organize (VRIO). The VRIO is an approaching that is capable to describe the competitive advantage significantly and the strate-gic plan in Times Higher Education Supplement (THES) is to obtain the ranking of the universities. The combination among fuzzy, VRIO framework, and THES is making a new concept in evaluating and arranging the strategic plan especially in the university.

Modern hand held devices such as smartphones and tablets have become demanding powerful. Wide range of commercial concepts and possibilities has come to limelight due to the extra features fitted in these devices. In particular, most smartphones and tablets regularly includes cameras, high-speed processors, GPS and internet access. With emerging technology, there are many applications, which allow the passing of real time information, and location based services. Mobile devices from various platforms gain much popularity and becoming more intelligent these days. In short, we are using them to accomplish our daily tasks. Hence, two application is developed and tested with cloud services integration.

Integrated connected car management system (ICCMS) technology focused on the car management. The system of ICCMS is the inevitable development trend of the next generation connected car diagnostic system, it is important for drivers to be connected of the cars details and as well convenient possibilities. This paper is based on a system development, to manage car internal states, requesting for service/towing and to operate in centralized database. At the end of the paper, the technical development prospect was discussed.

Location-based services are becoming a part of everyday life especially with the development of mobile communication. Examples of such applications include emergency services, navigation systems, tour planning or even finding the most convenient restaurant around. They have been playing an important role in providing necessary services efficiently with the use of Global Positioning System (GPS). When it comes to vehicle servicing, the problems faced by the users still remain especially in case of emergency breakdown mostly in finding the nearest services available. Consequently, this gives the encouragement to come out with a solution, which would help users ease their process of solving such problems. Vehicle Service Locator (VSL) allows user to locate the nearest tow trucks available, towing services and nearest vehicle service centers according to user's location data obtained via GPS and provides directions based on the user's current location.

Connected Vehicle Maintenance System is a system that is developed to assist car owners from facing problems especially when there is a vehicle breakdown. Vehicles in Malaysia do not have any available platform system that can provide assistance, diagnostic and telematics services to car owners. As vehicles are connected with various components of vehicle control unit, huge number of data are generated. Gathering this data has become a part of the connected vehicle research scope. However, car owners do not understand raw vehicle data unless their vehicles are brought to a service center. Thus, enabling in-vehicle system to interpret and monitor this data is an advancement of technology and provides a convenient driving environment for car owners.

Smart campus is an emerging concept in recent years that enables as a part of the smart city movements which leverages the use of information technology in building a comprehensive and sustainable living environment. Several campuses in Indonesia and abroad have already started the movement and integrate the colleges' data and provide a digital experience for the campus communities. In order to motivate the communities, especially in this case, the students to leverage IT tools and platforms to improve their learning experiences utilizing gamification and mobile technologies are seen to be the suitable combinations. Gamification as a concept to enhance user participation is also gaining a lot of tractions in recent years, including in the higher education sector. Many researches have shown that gamification is effective to improve motivations in core learning activities, to push students to be more proactive in gaining knowledge. This research is conducted to integrate a new mobile application to the existing smart campus platform in Universitas Multimedia Nusantara, Indonesia, while gamifying several student routines and activities to enhance the campus experience. The system is built as a hybrid mobile app and evaluated for its user acceptance by analyzing its users' perceived usefulness and perceived ease of use which are derived from the proposed technology acceptance model (TAM). Based on the feedbacks gathered from the users, the gamified application is perceived to be easy to use and useful to increase student's participation and to motivate the users to actively complete their campus routines.

Pneumonia is an acute infection that affects the lung tissue (alveoli) which can be caused by various microorganisms such as virus-es, fungi, and bacteria. Pneumonia was the second leading cause of death (13.2%) after diarrhea (17.2%) among under-fives. It shows that pneumonia is a disease that becomes a public health problem that contributes to the high mortality rate of children under five in Indonesia. In recent years, Bojonegoro city gives a large contribution for toddler pneumonia under five years of age in East Java district, Indonesia. In 2012, pneumonia sufferers reached 90.17% of the total number of the toddler and still many the next years. There are five risk factors for pneumonia include the number of children under five, the estimated number of patients, the number of sufferers, environmental factors and nutritional status. A spatial approach is needed to see the spreading of pneumonia vulnerability level in each sub-district in Bojonegoro. This approach can be used by the government as a supporting effort in controlling and preventing pneumonia which is more focused, efficient, and effective. This paper proposes a new approach to generate a vulnerability mapping of toddler pneumonia using hybrid genetic algorithm - K-means (GA-Kmeans) clustering algorithm accord-ing to five risk factors. K-means is a clustering algorithm that can produce data groupings based on several attributes well and quickly. However, there is a problem in the initialization stage of the initial random seeds from K-means, which is very difficult to reach an optimum global. The genetic algorithm is used to optimize initial seeds in the K-means algorithm. The vulnerability level of toddler pneumonia is classified into low, medium and high, then it is visualized into spatial mapping. The result of GA-Kmeans test iteration experiments produced best variance cluster 0.99 (almost 1) and determined high levels of vulnerability in 2016 are Kedungadem, Kepohbaru, Baureno, Kanor, Sumberrejo, Balen, Kapas, Bojonegoro, Dander and Ngasem sub-districts.

Urban agriculture is an agricultural activity around the city in the cultivation and processing of food. Urban agriculture generally produces fruits and vegetables that have a short shelf life. Urban agricultural activities include production, distribution and marketing of agricultural products produced. Agricultural production must be appropriate and efficient routes are distributed through. Errors in route selection can result in shipping, large operating costs and can reduce product quality, so that is necessary to choose the right route to distribute products that can reduce operational costs and maintain the quality of product until the destination. This research develops an application to determine the distribution route for urban agricultural products using genetic algorithms. This application can determine the optimal distribution route so that in the process of distribution of urban agricultural products can be done accurately and efficiently. Route optimization with genetic algorithms successfully optimizes the best route, can be seen from the fitness value that continues to increase each generation

Software development projects have relatively low success rates compared to other projects. One of the causes is human factors, lack of competence from the development team. The competency of the existing development team cannot meet project needs. Therefore, this research has analyzed the list of competencies that have an important role in software development team. The list of competencies analyzed was obtained from the results of literature review and evaluations from 17 practitioners in software development. From the literature review, 57 competency elements were obtained in the context of software developers, which will be further evaluated. Based on the evaluation of the importance level and frequency from practitioners that mapped in the importance-frequency matrix, 24 competency elements include as a top priority (in maintain or include (MI) quadrant) and 7 competency elements in possibly included (PI) quadrants. The results of this study are expected to be a reference for composing the right composition of the development team to achieve the success of the software development project.

The Learning Style Detection model in e-learning systems is experiencing rapid development. This development is characterized by the ex-istence of two learning style detection approaches namely automatic and conventional. The development of detection of automatic and con-ventional learning styles that exist today does not pay attention to the relationship of learning styles with prior knowledge. This is important to note because the style of learning is not static and tends to be dynamic depending on the topic of learning. This study builds the VARK MDP learning style detection model. It explores the relationship between learning styles with prior knowledge as evidenced by experiments on 32 learners. There are three steps taken: Measurement Prior Knowledge, Determine Prior Knowledge, Preference Learning Style. To evaluate this model we built detection scenarios with prior knowledge and compared with the results of interviews based on VARK learning style questionnaire. This study succeeded in building a model of measurement of prior knowledge that is more accurate than the previous model. Detection results also show that every learner does not only have one learning style and changes according to the topic.

Foreign Exchange (FX) is a conversion from one currency to another one and it can be seen as a discrete time series data. FX market has become one of the largest markets with more than trillions of dollars are traded every day. Therefore, the needs to capture FX transaction data pattern from historical data and used them to forecast future values have become a major issue. In this study, we tried to forecast the future values of FX transaction data using a relatively new moving average (MA) method, called as Weighted Exponential Moving Average (WEMA). Seven major currency pairs were being considered to be forecasted. Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) criteria were used to measure the accuracy level of the applied method. From experimental results on Phatsa framework, we found that WEMA can be used to predict future values of all major currency pairs used in this study.

With the fact that the tendency of startup business fail, it is necessary to understand factor influencing the startup success. It is understood that the lack of knowledge of how to start business creates uncertainty. The uncertainty may increase the likelihood of startup failure, which leads to economic loss. This paper aims to develop a predictive model of startup success in identifying factors and variables influencing startup success. This study also explores ways to compile and validate predefined startup prediction models in the startup business environment. Furthermore, the significance of factors and variables have also been identified to predict software startup success. The results showed that there were 18 significant factors influencing software startup success. These include the characteristics of the founders, quality of the startup organizations, support of resources, environmental conditions, performance of value creation processes, and effectiveness of competitive strategies. The result also showed that six variables contributed significantly (74.8%) in the prediction of software startup success.

Collaborative filtering with a traditional rating-based approach will generally use interaction records between users and systems to measure similarity, prediction, and generate recommendations. However, traditional rating based is not able to capture user preferences of different products. To overcome this issue, the ranking based approach, such as Borda method is used. The method takes advantage of rating data to determine the position of the product in the list of preferences as the basis for determining product points. The list of the preferences which is merely based on rating data, results in an insufficient accuracy of the recommendation. This paper, therefore, proposed a novel approach namely WP-Rank aggregation method which maximizes the use of rating data to generate product weight. The experimental results show that WP-Rank method is superior then Borda method with NDCG average value difference of 0.0220. However, WP-Rank requires longer running time with 0.0206 seconds lag from the Borda method.

This research aims to build desktop application called TabuCount with Borland Delphi 7.0 application in the Software Engineering subject. TabuCount helping other researchers or other students who used Tabu Search Algorithm like Travelling Salesman Problem to count each iteration step by step, and minimized human error for counting number. This research used Research and Development (R&D) design procedures through 5 steps, which are: (1) data classification, (2) data collecting, (3) design application, (4) field testing, and (5) building application. Based on the results of manual count with the TabuCount result, it was obtained the same result are valid and applicable. For an effective based on the research, TabuCount application were effective to use. Therefore, researchers who wanted to use Tabu Search Algorithm (Travel Salesman Problem) is suggested to use TabuCount as an alternative support application.

Changes in higher education cannot be avoided. The education system has changed from Education 3.0 to Education 4.0. Changes occur in various aspects of universities, both changes in technology and methods used, and changes in learning targets. Adaptation is necessary for the academic process and administrative processes in higher education. This is necessary in an effort to maintain the continuity of information technology (IT) quality management in higher education. Flexibility is an important consideration for universities. Flexibility can maintain sustainability and promises to increase value for universities. However, differences in university conditions can lead to differences in ability to adopt the concept of flexibility. There are inhibiting factors that cause resistance in adopting IT-based changes. These factors need to be managed using the right strategy. The good management is expected to minimize the cause of resistance so that the flexibility of IT management in universities can be increased. Currently the discussion of handling resistance focuses on companies and organizations in general. This paper focuses specifically on analyzing: factors that cause resistance to IT-based changes in universities are accompanied by adequate empirical evidence. The stages of the research conducted are identification of variables, building survey instruments, data collection, and testing variables. Research results that variable, financial support, human resources, and management support, simultaneously affecting IT-based change resistance in universities. Simultaneous influence percentage is 79%.

Fruits quality inspection is important task on agriculture industry. Automated inspection using machine and vision technology have been widely used for increasing accuracy and decreasing working cost. Convolutional Neural Network (CNN) is a type of deep learning that had a great success in large scale image and video recognition. In this research, we investigate the effect of different deep convolutional neural network architectures on its accuracy in strawberry grading system (quality inspection). We evaluate different types of existing deep CNN architectures such as AlexNet, MobileNet, GooLeNet, VGGNet, and Xception, and we compare them with two layers CNN architecture as our baseline. Here, we have done two experiments, the first is two classes strawberry classification and the second is four classes strawberry classification. Results show that VGGNet achieves the best accuracy, while GooLeNet achieves the most computational efficient architecture. The results are consistent on both two classes classification and four classes classification.

The recent advances of Internet of Things (IoT) technologies have changed the requirements of IoT device to not only provide basic sensing and communication services but also of executing more complex applications with different goals. These challenges have highlighted the need to provide high computation capability in IoT devices. However, common limited resources in IoT devices bring challenges to support application requirements as well as to deal with limited computation resources. To address with this problem, computation offloading can be applied. In this approach heavy computational tasks can be transferred and executed in the cloud computing service to get the result. However, sending heavy computational jobs along with the data to the cloud server are not always efficient, especially where the mobile environments where network performances may changes unpredictably. This paper proposes a prototype of smart offloading framework designed to work in IoT devices using the Fuzzy Multi Criteria Decision Mak-ing as the decision tool. The decision whether the job execution will be done in the IoT device itself or being uploaded to the cloud computing server is done by considering internal and external factors such as current network conditions. The smart offloading framework prototype has been developed and tested in a real IoT device. The experiment results showed that the smart offloading approach can improve the performance of applications running in IoT device in deciding optimal location of the jobs execution in dynamic situations with good results.

Computer supported cooperative work (CSCW) is one of the key factors to increase company's productivity. Hence, there are numerous researches trying to increase the efficiency of CSCW implementation to achieve a better collaboration between employees. The recent trend of using 3D content demand industry to develop an efficient and effective workflow to create numerous high-quality 3D content. Hence, implementing CSCW in 3D content authoring is a crucial factor in related industry such as game and movie industry. In this paper, we in-vestigate the state of collaborative work in 3D content authoring. We focus our observation in CSCW features currently available 3D con-tent authoring application and game engines. Additionally, we observe recent researches related to developing CSCW application in 3D authoring. The objective of this research is to give a thorough view regarding the state of CSCW in 3D content authoring industry and the possibility of future research and development.

In 3D content creation, developing a terrain may be time consuming due to most of current 3D applications require a vast terrain. In this paper, we introduce a collaborative terrain editor that allows multiple users to collaborate in real time. The objective of the application is to increase the productivity in developing a vast terrain in 3D environment. The application is based on a client server architecture where each editor applications are connected to a single server that collects and distributes the editing process done by each user. We perform our test in a local area network environment with multiple clients connected to a server and observe the performance and the usability of the software. The test results show the system is capable to perform real-time collaborative terrain editing with insignificant delay. Moreover, most of our users agree that using the system may increase the speed in 3D terrain creation.

In the research field of signal processing by using wavelet method there are some factors affecting the accuracy of recognition such as the selection of the sub-band parameter, the selection of suitable mother wavelet or coefficient, and the determination of decomposition level. This paper presents a comparative study of three wavelet-based sub-bands (WBSB) combined with the moving average energy (MAE) features for classification of Indonesian stop consonants in consonant-vowel (CV) context. Three different feature sets used in this study are the moving average energy of each different wavelet sub-band using mother wavelet of daubechies2. The first feature set is the MAE taken from standard wavelet packet (WP) sub-band at the 4th level of decomposition denoted as WBSB. Whereas the second and third feature sets are the MAE taken from WP which the sub-band is selected based on the previous research denoted as WBSB1 and WBSB2. Each CVC syllable is segmented at a certain length to form a CV unit. For the classification of the stops, two different classifiers were used, based on multi-layer perceptron and random forest. The experimental result showed that the performance rank of feature extraction method were WBSB, WBSB1, and WBSSB2, respectively

The aim of this study is to extract entity locations on crime news in Indonesian online newspapers and to tag the locations into a map. The methods used in this study are rule-based algorithm, for identifying and extracting entity location of the crime, and SVM (Support Vector Machine) algorithm, for classifying which sentence containing the location of the crime. Every sentences containing criminals location was included in geocoding process so it could be mapped into a digital map. The accuracy of identifying the entity location by using rule-based algorithm is 96.2%. SVM model that has the best accuracy in classifying sentences that contains entity scene of the crime is Radial kernel whose accuracy is 95.77%.

The development of utilization of outdoor wireless access point devices has progressed very rapidly. Mitigation and control of the use of the frequency spectrum are very important to do so that the use of the frequency spectrum can run in accordance with government regulations. This cannot be separated from the technique of localizing the use of wireless devices. There are various localization techniques with various methods and levels of accuracy but generally, these techniques are applied indoors. Localization techniques are generally used to find the location of wireless users, not to search for access point localization. In this paper the distance estimation system from the fingerprint-based outdoor wireless access point localization system is discussed, and wireless devices working in 2.4 GHz. The distance estimation system uses regression method, and this paper aims to prove that 3rd order regression polynomial is the right regression model used for fingerprint-based wireless access point localization. Previously this technique was applied at a distance of 0-100 meters, so this paper confirms this technique is applied at a distance of 0-1000 meters. The fingerprint is carried out in the range of 0 to 1000 meters and is divided into eleven measurement points. DataPoint consists of received signal strength (RSSfnd) and the distance of the finder to an access point that is being targeted (Dfnd). DataPoint from the fingerprinting process is analyzed by regression method and based on the results of the trendline and R2 shows that in the range of 0-1000 meters the regression method that is right to use for distance estimation system is the 3rd order polynomial regression.

The Spectrophotometric and Colorimetric method are the most prominent method that widely used for in the various scientific fields such as physics, materials science, chemistry, biochemistry, and molecular biology. Those methods were based on the application of Beer and Lambert Law. However, the current available instrument used for measurement are mostly take use the combination of the light source, grating disc and the silicon photocell. Unfortunately, the combination of those components causes the price to increase. The proposed system was developed with the low cost, which mainly utilizing the Light Emitting Diodes (LED) as the light sources and RGB Sensor. There are found two types of LED, which have the different wavelength selected for each solution and the preferred method to be used for the measurement: 410nm LED for Yellowish Solution and 570nm LED for reddish solution. The USB Serial (Through USB Cable) and Wireless Bluetooth (The Bluetooth Module) were both the option for the communication method to control the system and see the results. The measurement range is 0.02 to 30.0 ppm. The colorimeter was successfully applied for determination of nitrate and fluoride concentrations with the average R correlation of 0.99 when compared to those obtained by using a spectropho-tometer (Optizen POP QX, Mecasys, South Korea) used for solution concentration measurement. The material cost was less than $75 USD.

This paper describes the development of a portable instrument for measuring the conductivity and temperature of water. The system consisted of thick film sensors fabricated on alumina substrates using screen printing technique, and Arduino microcontroller to process and display the measured water parameters. The conductivity sensor was designed in a four-electrode configuration using AgPd conductor with a cell constant of 0.2 cm-1, whereas the temperature sensor was screen printed using RuO2 paste. The system has been tested to measure the conductivity of various water samples, including river and drinking water, and comparison with a commercial system demonstrated an error of less than 2.4 %. The maximum conductivity value that could be measured was under 2000 microS/cm, which makes the system suitable for monitoring of general drinking water and domestically polluted river water.

As the number of computer users increases, numerous content has been generated by them. Machine learning as one of the main direction of natural language processing, allows computer systems to extract various information from the generated content. Processing results deter-mine the sentiments of the text to extract the author's emotional evaluation that is expressed in the text. The aim of the project was to develop the Sentiment Analyzing system by using Machine Learning algorithms on cloud-based system. The paper describes the development pro-cess of Sentiment Analyzing System in English language. Two Machine Learning algorithms, SVM and Naïve Bayes classifier, have been inspected and Cloud computing used to develop and publish web application. The testing results demonstrate the accuracy of the work in proposed method.

The K-Means clustering algorithm is an unsupervised data mining technique and it has also been used widely to solve the problem in real life. This paper addresses a design analysis of the algorithm of K-Means that is implemented in the field programmable gate arrays (FPGAs). These devices are integrated circuits and have a hardware platform and applied in many implementations. The K-Means clustering algorithm has a simple method that is choosing objects from data to become the center point or centroid and then assign each object to the cluster which is nearest and update the cluster means. The approach of the software can also be implemented in the hardware, but both of them have a different method in programming.

Extreme event such as strong shock resulting from violation of the hole or a large object on the road can cause damage to vehicle components. As such, the study needs to be done to address the behavior of failure data using the fatigue life and reliability characteristics of extreme fatigue life failure statistical approach. The study can also be done by developing a characterization of life distributions based data and configuration best match load to allow a generalized prediction. The research involves testing the fatigue life and cyclic strain fatigue life data generation using Monte Carlo simulation based on the parameters of probabilistic stress cycle curve. Features for all parametric distributions were analyzed by the method of maximum likelihood estimation (MLE) for the generalized extreme value distributions. Assess the suitability of the life distribution for the reliability of extreme fatigue life can be seen through a probability density function. This study found that the developed method capable of predicting the relationship between the load configuration and shape of the distribution of a component failure studied. This approach can contribute to reduced time of experimental testing which is emphasis in the production process components. This implication provides a particularly significant impact on the development of the automotive industry and enhance the manufacturing sector.

Finite element analysis (FEA) is extensively used in the engineering field, especially in structural engineering. FEA is a numerical method used for solving engineering problems that involve complicated geometries, various loading conditions and material properties. A fatigue life assessment of an engineering component is necessary to predict the life span of this component before failure. Automobile alloy wheel is an engineering component that is exposed to fatigue failure in services. Therefore, this study aims to determine the critical area for crack initiation on automobile alloy wheel and to simulate and analyse the fatigue life of an automobile alloy wheel design that is fabricated from different types of materials. The automobile alloy wheel design was modelled using computer-aided design and analysed using commercial finite element software. The automobile alloy wheel was modelled based on a 1200 cc national automobile. Three types of materials, namely, titanium, aluminium and magnesium alloys, were used in this study. A critical part of a steering knuckle could be identified by conducting a stress analysis, and the fatigue life of the automobile alloy wheel could be predicted on the basis of applied loads. Results showed that fatigue life is significantly influenced by the types of material used in a simulation.

Laminated fibre reinforced composite can be subjected to periodic loading that can cause parametric instability in structures. This paper is to report on the parametric instability analysis of unsymmetric laminated composite plate. Using finite element method, the Mathieu-Hill equation that describes the parametric instability behavior of composite plate was derived based on the higher order shear deformation theory of composite and the equation was solved using the Bolotin's method. Compared to similar symmetric composite, the anisotropic nature of the composite has caused the composite to be more unstable by shifting the instability region to the left while narrowing the region at the same time. However, the effects of parameters such as the orientation angle and the aspect ratio are similar to the effect of the same parameters on the parametric instability of symmetric composites.

The design of a microbubble generator that has high efficiency and good performance is still a challenge today, especially for a large scale application. In this study, CFD simulation based on the transient operation was used to predict the characteristics of a fluid flow as a reference in the design process. To analyze the performance of the swirl flow microbubble generator, particle image velocimetry (PIV) was used to characterize the dimension and distribution of microbubbles. Based on the simulation results, CFD was able to visualize the mixing process and the fluid characteristics of the gas-liquid flow in a swirl flow microbubble generator. Air self-suction mechanism in a microbubble generator nozzle was successfully formed by a negative pressure in the central axis area of the nozzle due to a swirl flow of water. It shows that a swirl flow microbubble generator can work efficiently and doesn't need any other devices to deliver air into the system. Based on the PIV measurement, the radius of microbubbles was successfully formed and averaged of 25 µm for both air mass flow rate of 0.25 l/min and 1 l/min. However, the smaller the mass flow rate of air, the more the number of microbubbles generated.

This paper presents an optimization of permanent magnet synchronous machines (PMSMs) using Analytical Sub-domain Model together with Differential Evolution Algorithm (ASDEA). A three-phase, 6-slot/4-pole, surface-mounted PMSM is selected in the design with initial motor parameters which are determined from the sizing equations. Five motor parameters are to be optimized i.e. magnet thickness, airgap length, slot-opening width, magnet arc, and stator inner radius. Four objective functions are chosen i.e. to have lowest total harmonic distortions in the induced back-emf, lowest cogging torque, highest output torque and highest efficiency. Results show a good agreement between the analytical method and finite element analysis (FEA). The optimization of 6-slot/4-pole PMSM is further analyzed by comparing with other optimization algorithms i.e. Analytical Sub-domain with Genetic Algorithm (ASGA), and Analytical Sub-domain with Particle Swarm Optimization (ASPSO). It is observed that ASPSO has the fastest computing time compared to ASGA and ASDEA. Whereas ASDEA is approximately 50% faster than ASGA. The design work for PMSMs can potentially become faster without compromising the accuracy. While repetitive changes in motor parameters in finite element modeling could be avoided after applying this Analytical Sub-domain with Differential Evolution Algorithm.

Unmanned Aerial Vehicle (UAV) industry shows rapid growth in producing sophisticated UAVs for civilian use in recent years. With the invention of private UAV networks, these UAVs can now be used in a UAV network owned by a single organization or a person. For example, a private UAV network gives the ability for a farmer to monitor the farm from one location or to locate people in a natural disaster-stricken area. The wireless communication paths between UAVs and the ground station in a private UAV network play a vital role in transmitting video to the ground station. It is also necessary that the control signals are sent over the private UAV network to the UAVs to control their movements. Since the control signals cannot be delayed and videos received at the ground station should be of high quality, separate channels should be created for user data and control data traffic. In this paper, we present a channel allocation approach for private UAV networks to transfer control data as well as user data. For high quality video transmission, we recommend bonding of basic video channels. This new channelization also avoids congestion and packet drop issues of video and control data traffic in private UAV networks.

The electronic data has to be made because some reasons which are ease of use, sharing, analyzing, and ability of storing a large volume in a small space. The term of the database is a collection of the files that is used to store data. In the application of the software it can be easy to understand but in the aspect of the hardware, the function is same with how the database in software do. The applications of database have touched and spread all aspects of information. Mean-time, there is another term which is called a database management system (DBMS). A DBMS is a C program that allow the user to manage a large database efficiently and stored at a long time. There are examples of the DBMS e.g. Oracle, DB2, Sybase, SQL Server, etc. The main functions of the database are create/store, search, update, etc.

Robots are closer than ever to leap from the defined world navigation to the undefined world navigation. For that, it is crucial to have flexible and alternative platforms to enhance mobile self-made robots to the next level. In this project, a vision based navigation robot is developed using Microsoft Windows-based software as an alternative to the most common Linux based Robot Operating System (ROS). Currently, almost all self-made robots with advanced functionalities use the ROS. But what lacks in Linux based system is flexibility for beginners. Developing a mobile robot based on the Windows platform was not possible a few years ago, but with the development of Single Board Computers (SBC), now finally it is possible to mobilize the windows. It's a marriage between windows flexibility to robotic mobility. The final result of this research is an autonomous robot which is able to navigate through the environment by avoiding obstacles in a dynamic environment using 3D vision. The solution comprises occupancy grid generation by using depth image of Kinect as main input feed combined with data gathered by an Inertial Measurement Unit (IMU). Data then further processed real-time to identify the current position of the robot. Also, by using Fuzzy Logic approach, dynamic obstacle avoidance and navigation is achieved. The entire application was developed by LabVIEW and it is installed inside LattePanda single board computer. LattePanda is playing a crucial role here as the central processing unit for all the functions. The robot is developed as a first step for a flexible test platform which can be further improved to achieve flawless undefined environment navigation through tweaking algorithms and introducing more sensory data.

Biofilter is one of a method to remove odorous gas, because of inexpensive, easy, and can remove the odor perfectly. This research are to determine the best of physical characteristic of bokashi compost, husk charcoal, and wood charcoal toward the adsorption of ammonia and compositions of such materials. The research is used Ammonia solution with concentration 5%. The physical characteristic of packing material was conducted to assess certain parameter such as moisture content, density, porosity, pressure drops, water holding capacity (WHC), pollutant holding capacity (PHC). Data were analyzed by ANOVA statistic test with error value 5% and continued with Duncan Test. From the physical experiment, by performance ranking shows husk charcoal is better than wood charcoal, and compost. Husk carbon shows the highest porosity, availability in hold water and air, PHC, and pressure drops. Husk carbon can adsorb 1.18x10-3 g-N/g bk of ammonia. Treatment of material composition bokashi, husk charcoal, and wood charcoal with a ratio of 2:1:2 (K212) showed that K212 is the best composition in ammonia absorption. This is evident from the water content of K212, lower pressure drops, PHC of 1.8x10-4 g-N / g bk.