Proceeding of the Electrical Engineering Computer Science and Informatics
Not a member yet
649 research outputs found
Sort by
Analysis of Statement Branch and Loop Coverage in Software Testing with Genetic Algorithm
Get PDFSoftware testing is one important aspect of the software development process. About 50% of the time and cost in the software development process used for software testing process. There are two methods of software testing, black-box testing and white-box testing. This research using white-box testing. Software testing can be done manually or automatically. Based on research conducted, genetic algorithm has been widely implemented in software testing, such as test data generator. The purpose of this study is to apply a genetic algorithm in software testing and comparing the results with manual testing, automated, and automated with genetic algorithm. The test parameters are coverage measurements (statement, branch and loop coverage) and the time of testing. The conclusion of this study is automated testing with genetic algorithm requires fewer time and test cases to achieve coverage of 100
Combining Deep Belief Networks and Bidirectional Long Short-Term Memory
Get PDFThis paper proposes a new combination of Deep Belief Networks (DBN) and Bidirectional Long Short-Term Memory (Bi-LSTM) for Sleep Stage Classification. Tests were performed using sleep stages of 25 patients with sleep disorders. The recording comes from electroencephalography (EEG), electromyography (EMG), and electrooculography (EOG) represented in signal form. All three of these signals processed and extracted to produce 28 features. The next stage, DBN Bi-LSTM is applied. The analysis of this combination compared with the DBN, DBN HMM (Hidden Markov Models), and Bi-LSTM. The results obtained that DBN Bi-LSTM is the best based on precision, recall, and F1 score
A Moving Objects Detection in Underwater Video Using Subtraction of the Background Model
Get PDFThis paper proposes a method for detecting moving objects on an underwater video. Video obtained using an underwater camera to capture the environmental conditions of the area. This research is the initial stage of the underwater surveillance system. Underwater surveillance system enables objects passing can be recognized shapes, types, and its behavior. The detection method used in this research is a subtraction between the current frames with the background modeling results. Underwater video retrieval has a high level of difficulty because the background is always changing either due to a change the intensity and the movement of water currents. Therefore, it needs to be made an appropriate background model to address this problem. Modeling of the background on this research using adaptive modeling method, where the intensity of the background pixels is updated based on inference of the background intensity before. If the intensity of the pixels changed drastically beyond the allowed threshold value, the pixel is considered as the pixels of the object and the pixel values of the background model are updated based on this pixel value. The effectiveness of the proposed method is expressed with the value of recall and precision. The average recall value of the two videos is 83% and the value of its precision is 67.5%
Honey Yield Prediction Using Tsukamoto Fuzzy Inference System
Get PDFHoney is a natural product of bee. Since ancient times, honey has been known by humans as a source of natural food and also for traditional medicine. There are so many beneficial of honey, make people trying to do honeybee cultivate as a business solution to increase their income. However, to cultivate honey bees is not easy. Special knowledge is required on honey bee cultivation and capital is fairly large. In order for beekeepers not to lose from honey sales business, beekeepers should be able to estimate the honey yield accurately. Predicted yield of honey is used as a material consideration and help determine the decision in honey bee cultivation. This study provides a solution for prediction of honey yield type Apis Cerana with the main food of Calliandra flowers accurately. The method used in this research is Tsukamoto's fuzzy inference system (FIS) method. There are 3 input fuzzy used in this study, namely : Rainfall, number of box, and number of flower trees. The three fuzzy inputs are the determinants of the honey yield. The representation model used in the research is Trapezoid with fuzzy rules of 125 rules. While the test data in this research are rainfall and honey yield data for 21 years. The results of this study showed that the prediction of honey yield using FIS Tsukamoto closed the real honey yield with RMSE value of 9.44933860119277
Big Data Management Prototype Development for Analysis Various of Data
Get PDFThe phenomenon of big data is currently agrowing topic in the world of information technology. From someof the literature mentioned that manage big data can createsignificant value for the world economy, improving productivityand competitiveness of enterprises and the public sector as wellas creating a large economic surplus for consumers. However,from some of the information obtained, big data is still not widelyapplied in the company or organization. This study aimed toexplore more information about the big data and proceed withmaking an application prototype big data management. Toexperiment with big data storage that is database, this researchuse NoSQL database technology that can map the needs of bothstructured and unstructured. And this research will be carriedout migration of Relational Database (RDBMS) into the databaseMongoDB. Prototype will be create with the object of study isstructured and unstructured data. The expected result of thisresearch is a model or prototype of big data management thatcan help organizations and companies (especially education) tomake decisions based on various types of data
Decision Support System for Heart Disease Diagnosing Using K-NN Algorithm
Get PDFHeart disease is a notoriously dangerous disease whichpossibly causing the death. An electrocardiogram (ECG) is used fora diagnosis of the disease. It is often, however, a fault diagnosis by adoctor misleads to inappropriate treatment, which increases a riskof death. This present work implements k-nearest neighbor (K-NN)on ECG data to get a better interpretation which expected to help adecision making in the diagnosis. For experiment, we use an ECGdata from MIT BIH and zoom in on classification of three classes;normal, myocardial infarction and others. We use a single decisionthreshold to evaluate the validity of the experiment. The resultshows an accuracy up to 87% with a value of K =
Computer Anxiety and Computer Attitude towards Computer Self Efficacy (CSE) Polsri Telecommunication Engineering Student on Writing the Final Report
Get PDFVarious attitudes emerged and shown byindividuals for the presence of computer. Although manybenefits are felt by the computer, but there are some peoplewho feel anxious with the computer (computer anxiety).Computer attitude showed no reaction or behave the computerby pleasure or displeasure against the computer. Thephenomenon that arises is computer anxiety and computerattitude can affect a person's expertise in the use or operate thecomputer. This study aims to examine how the effect ofcomputer anxiety and computer attitude towards computer selfefficacy on the 6th semester student telecommunicationsengineering POLSRI. Samples taken in this study were allstudents of telecommunication engineering 6th semesterPOLSRI totaling 89 people. The results showed computeranxiety and computer attitude not affect the computer selfefficacy6th semester student of telecommunicationsengineering POLSRI. This is due because the 6th semesterstudent of telecommunications engineering POLSRI havepositive feelings to learn the computer either by themselves orthrough a learning courses.They also realize that the computerprovides many benefits. With the computer, the informationcan be obtained more quickly and efficiently. Computer is anecessity, can enhance human life, and was instrumental ineducation and employment
Using Q-Learning for Recommend Learning Object on e-Learning System
Get PDFThis paper discusses the software development used for theselection of instructional materials based on an appropriatelearning style. Q-Learning is used to estimate state-actionlearners using Q-Learning. It is conducted by calculating thevalue of the action done to reward students in selectingteaching materials, so the pattern matches are found. Thisstudy uses the Felder Silverman Learning Style Model(FSLSM). Tests showed that the system is able to perform theselection (action) of instructional materials for teachingmaterials that match learning styles. Q-Learning Lessons canbe applied to optimize the selection of instructional materialswith minimal effor
The Elimination of Overshoot Curve Response of Closed Loop in Proportional Integral (PI) Controller
Get PDFMost operators in industry use trial and error methodin determining the parameter in PID controller. This way isquite dangerous because it cannot predict what will happen inthe next process. For this, we need a method that can adjust thechanging of parameter in process, and simultaneously retune theparameter of controller automatically. Ziegler-Nichols method, amethod for setting parameter of PID controller, can be use foreliminating the oscillation and reducing overshoot curve in aprocess. This method is common yet, but it offers simpleprocedure but produce quick and accurate result. This methodanalyzes the curve of a process. It is done when oscillation andovershoot projected onto x-axis (time) occurred. The parametersresulted from this analysis among others are: critical gain (Kpu),time/period oscillation (Tosc), static gain (K), proportional gain(Kp), integral time Ti, and differential time Td. These parameterswill be tuned to PID controller using Ziegler-Nichols method.The overshoot (Mp) curve response of closed loop in thissimulation is 0 %. It will save energy and time beside that thestability of system can be maintained
Development of Fuzzy Logic Based Temperature Controller for Dialysate Preparation System
Get PDFPreparing the dialysate temperature to the desiredlevel is a complicated task, since it has a large degree of time delayand nonlinear behaviour. In this work, embedded system fordialysate temperature controller was developed. It is based onimplementation of fuzzy logic controller software on STM32 F4development kit which consits of ARM Cortex M4Fmicrocontroller. The dialysate temperature was controlled byvarying the firing angle of the triac which is connected to theheater. The K-type thermocouple which is connected to AD595CQwas read and compared with the desire dialysate temperature.The fuzzification process was conducted prior to activate the fuzzyinference process. The power of the heater was adjusted based onthe output of the inference process until the desired watertemperatur was achieved. The system had been sucessfullydemonstrated for controlling the dialysate temperature