163 research outputs found
A hybrid feature selection on AIRS method for identifying breast cancer diseases
No full textBreast cancer may cause a death due to the late diagnosis. A cheap and accurate tool for early detection of this disease is essential to prevent fatal incidence. In general, the cheap and less invasive method to diagnose the disease could be done by biopsy using fine needle aspirates from breast tissue. However, rapid and accurate identification of the cancer cell pattern from the cell biopsy is still challenging task. This diagnostic tool can be developed using machine learning as a classification problem. The performance of the classifier depends on the interrelationship between sample sizes, some features, and classifier complexity. Thus, the removal of some irrelevant features may increase classification accuracy. In this study, a new hybrid feature selection fast correlation based feature (FCBF) and information gain (IG) was used to select features on identifying breast cancer using AIRS algorithm. The results of 10 times the crossing (CF) of our validation on various AIRS seeds indicate that the proposed method can achieve the best performance with accuracy =0.9797 and AUC=0.9777 at k=6 and seed=50
HOG Feature Extraction and KNN Classification for Detecting Vehicle in The Highway
No full textAutonomous car is a vehicle that can guide itself without human intervention. Various types of rudderless vehicles are being developed. Future systems where computers take over the art of driving. The problem is prior to being attention in an autonomous car for obtaining the high safety. Autonomous car need early warning system to avoid accidents in front of the car, especially the system can be used in the Highway location. In this paper, we propose a vision-based vehicle detection system for Autonomous car. Our detection algorithm consists of three main components: HOG feature extraction, KNN classifier, and vehicle detection. Feature extraction has been used to recognize an object such as cars. In this case, we use HOG feature extraction to detect as a car or non-car. We use the KNN algorithm to classify. KNN Classification in previous studies had quite good results. Car detected by matching about trining data with testing data. Trining data created by extract HOG feature from image 304 x 240 pixels. The system will produce a classification between car or non-car
Penjadwalan Kapal Penyeberangan Menggunakan Algoritma Genetika
No full textAbstrakPenyusunan penjadwalan kapal penyeberangan di Pelabuhan Ketapang – Gilimanuk sangatlah penting agar para penumpang pengguna jasa kapal laut mendapatkan pelayanan yang maksimal. Karena pembuatan penjadwalan masih dibuat secara manual sehingga memungkinkan ada nama kapal yang sama dalam sehari yang beroperasi dan tidak adanya keadilan pada pembagian porsi masing-masing kapal. Untuk mengatasi permasalahan tersebut diperlukan suatu sistem komputerisasi penjadwalan kapal guna mempercepat pengaturan jadwal pemberangkatan kapal penyeberangan Ketapang – Gilimanuk. Penerapan metode algoritma genetika dalam permasalahan penjadwalan kapal mampu menghasilkan solusi yang baik dengan menggunakan representasi kromosom permutasi bilangan integer, metode crossover menggunakan one cut-point crossover, mutasi menggunakan reciprocal exchange mutation, dan seleksi menggunakan elitism selection. Dari pengujian parameter didapat hasil yaitu antara lain ukuran populasi sebesar 180, banyaknya generasi 200, serta kombinasi crossover rate=0,6 dan mutation rate=0,4.Kata kunci: kapal, penjadwalan, algoritma genetika.AbstractArrangement schedule for dispatching ships in Ketapang – Gilimanuk Harbor is strongly important to make all the passengers get the best service. As the schedule arrangement is made manually, it is possible that there are ships with same name work in a day; another problem is there is no equality in distributing the portion for each ships. To solve that kind of problems, dispatching schedule using computerization is needed in order to make the dispatching schedule of the ships faster in Ketapang – Gilimanuk Harbor. The use of genetic algorithm method in dispatching schedule of ships produces a good solution with using representative of numeral integer chromosome permutation, crossover method using one cut-point crossover, mutation using reciprocal exchange mutation, and selection using elitism selection. From parameter test’s result, there are some outcomes such as 180 population size, 200 the amount of generation, and also the combination of crossover rate=0,6 and mutation rate=0,4.Keywords: ship, schedule, genetic algorithm
Rainfall Prediction in Tengger, Indonesia Using Hybrid Tsukamoto FIS and Genetic Algorithm Method
No full textCountries with a tropical climate, such as Indonesia, are highly dependent on rainfall prediction for many sectors, such as agriculture, aviation, and shipping. Rainfall has now become increasingly unpredictable due to climate change and this phenomenon also affects Indonesia. Therefore, a robust approach is required for more accurate rainfall prediction. The Tsukamoto Fuzzy Inference System (FIS) is one of the algorithms that can be used for prediction problems, but if its membership functions are not specified properly, the prediction error is still high. To improve the results, the boundaries of the membership functions can be adjusted automatically by using a genetic algorithm. The proposed genetic algorithm employs two selection processes. The first one uses the Roulette wheel method to select parents, while the second one uses the elitism method to select chromosomes for the next generation. Based on this approach, a rainfall prediction experiment was conducted for Tengger, Indonesia using historical rainfall data for ten-year periods. The proposed method generated root mean square errors (RMSE) of 6.78 and 6.63 for the areas of Tosari and Tutur respectively. These results are better compared with the results using Tsukamoto FIS and the Generalized Space Time Autoregressive (GSTAR) model from previous studies
New insight in cervical cancer diagnosis using convolution neural network architecture
No full textThe Pap smear is a screening method for early cervical cancer diagnosis. The selection of the right optimizer in the convolutional neural network (CNN) model is key to the success of the CNN in image classification, including the classification of cervical cancer Pap smear images. In this study, stochastic gradient descent (SGD), root mean square propagation (RMSprop), Adam, AdaGrad, AdaDelta, Adamax, and Nadam optimizers were used to classify cervical cancer Pap smear images from the SipakMed dataset. Resnet-18, Resnet-34, and VGG-16 are the CNN architectures used in this study, and each architecture uses a transfer-learning model. Based on the test results, we conclude that the transfer learning model performs better on all CNNs and optimization techniques and that in the transfer learning model, the optimization has little influence on the training of the model. Adamax, with accuracy values of 72.8% and 66.8%, had the best accuracy for the VGG-16 and Resnet-18 architectures, respectively. Resnet-34 had 54.0%. This is 0.034% lower than Nadam. Overall, Adamax is a suitable optimizer for CNN in cervical cancer classification on Resnet-18, Resnet-34, and VGG-16 architectures. This study provides new insights into the configuration of CNN models for Pap smear image analysis
Pengelompokan Data Hasil Tes Kepribadian 16pf Sopir Bus Menggunakan Algoritma Genetika
No full textAbstrakTes kepribadian merupakan suatu metode tes yang disusun untuk mendeskripsikan bagaimana kecenderungan seseorang dalam bertingkah laku maupun berpikir. Tes kepribadian sebenarnya hanya dapat dideskripsikan secara kualitatif karena sebenarnya kepribadian tidak dapat diukur. Namun, untuk membantu menjelaskan kepribadian seseorang dapat menggunakan bantuan angka sehingga hasil dari tes tersebut dapat di deskripsikan ke dalam bentuk kualitatif. Dalam penelitian sebelumnya, pengelompokan hasil data tes kepribadian 16PF dilakukan dengan metode K – Means Clustering yang digabung dengan metode Silhouette Coefficient. Penelitian tersebut memiliki hasil maksimum SI (Silhouette Index) hingga 0.4341. Dalam penelitian kali ini, metode yang dapat digunakan untuk mengelompokkan dan menghitung seluruh data serta atribut yang diperoleh menggunakan Algoritma Genetika. Tahapan untuk mengelompokkan data menggunakan metode yang sama seperti penelitian sebelumnya yaitu K – Means Clustering dan untuk menghitung cluster diperlukan representasi kromosom agar dapat membangkitkan nilai Centroid untuk perhitungan Silhouette Coefficient. Representasi kromosom yang digunakan adalah real code genetic algorithm dimana representasi tersebut dibangkitkan secara random dengan interval tertentu. Dari pengujian yang dilakukan, sistem mampu memberikan nilai SI terbaik pada jumlah populasi 40, jumlah generasi 15, kombinasi cr 0.7 dan mr 0.3. Algoritma genetika mampu memberikan solusi optimal dibandingkan dengan penelitian sebelumnya dimana dengan jumlah data yang sama menghasilkan nilai SI yang lebih baik.Kata Kunci: Algoritma genetika, Personality Factor, Clustering, K – Means, Silhouette Coefficient.AbstractAbstractPersonality Test is a test method developed to describe how the tendency of a person\u27s behavior and thinking. Actually, personality tests can only be described qualitatively because actual personality cannot be measured. However, figures can be used to help explaining an individual’s personality, thus test results could be described into qualitative terms. In previous research, grouping data results 16PF personality test was conducted using K - Means Clustering combined with Silhouette Coefficient methods. The study has a maximum performance in terms of SI (silhouette index) of 0.4341. In the present study, the method can be used to classify, count all the data and attributes that are obtained using Genetic Algorithms. Stages for classifying data using the same method as previous research, that K - Means Clustering and to calculate cluster, required the representation of chromosomes in order to generate value of Centroid for the calculation Silhouette Coefficient. Chromosome representation used is real code genetic algorithm which is representations generated randomly with a certain interval. From the tests, systems are able to provide the best SI values in populations of 40, the number of generations 15, combination of cr mr are 0.7 and 0.3. Genetic algorithms are able to provide optimal solutions compared to a previous study in which the same amount of data to produce better value SI.Keywords: Genetic algorithms, Personality Factor, Clustering, K – Means, Silhouette Coefficient
Optimisation of integrated multi-period production planning and scheduling problems in flexible manufacturing systems (FMSs) using hybrid genetic algorithms /
No full text1 ethesis (vii, 235 pages) :illustrations.Includes bibliographical references (pages 152-163)The productivity and efficiency of FMSs are strongly determined by their production planning and scheduling. The part type selection and the machine loading problems are strongly related problems in production planning of FMSs. Since the FMSs have technological constraints such as limited number of machines and tool magazines capacity of each machine, the production must often be done in batches. The part type selection deals with selection of subset of part types to be produced immediately in current batch.The allocation of operations and their machines for the selected part types and loading appropriate tools to the machines are addressed by the machine loading. The order of operations and their starting time on each machine is addressed by the scheduling. The integration of production planning and scheduling problems for several production periodsis needed to achieve a global optimum solution
Solving the integrated problems requires a powerful method to deal with a large searchspace. Genetic Algorithms (GAs) that have been proven as a robust meta-heuristic method to solve various complex problems with a huge search space were chosen to address the problems. The mathematical formulation for the integrated problems is developed to provide the foundation of mathematical computations and constraint satisfactions at theprogramming stage of the hybrid GA (HGA). The HGA is developed by hybridising thereal-coded genetic algorithm (RCGA) and the variable neighbourhood search (VNS). As the reproduction operators of the RCGA are designed to explore a large search space, theVNS is employed to enhance the power of the HGA exploiting local optimal areas. The chromosome representation of the HGA is designed to produce only feasible solutions that minimise the computational time required by the HGA during generations. A strategy to maintain population diversity is implemented to avoid a premature convergence. The strategy is supported by mechanism to adaptively adjust crossover and mutation rates.Thesis (PhDEngineering)--University of South Australia, 2013
- …
