Hasil Pencarian  ::  Simpan CSV :: Kembali

"ABSTRAKKanker payudara merupakan penyebab kematian pertama akibat kanker pada wanita. Pria juga dapat terkena kanker payudara. Penanganan kanker payudara terdiri dari operasi, terapi radiasi, dan terapi sistemik yang menggunakan obat. World Health Organization WHO telah mendaftarkan tiga puluh cytotoxic dan obat antikanker untuk mencegah dan mengurangi kanker payudara. Para ilmuwan sudah berusaha untuk menemukan obat lain untuk membantu orang yang terkena kanker payudara. Oleh karena itu, desain obat menjadi penting dalam menemukan obat baru yang potensial untuk menangani kanker payudara. Pada skripsi ini diajukan pendekatan multiple linear regression MLR menggunakan metode quantitative structure activity relationship QSAR untuk memodelkan desain obat kanker payudara dengan pemilihan variabel menggunakan metode algoritma genetika GA . Data yang diperoleh dari bank protein umum lebih sedikit dibandingkan banyaknya variabel, yang menyebabkan asumsi Analisis MLR gagal dan menyebabkan multikolinearitas. Model QSAR diragukan saat terjadi multikolinearitas. Algoritma Genetika GA diimplementasikan untuk menghilangkan multikolinearitas. GA bertindak sebagai penyeleksi variabel untuk mencari variabel yang signifikan dan membantu mendapatkan model QSAR yang paling cocok. Hasil eksperimen menunjukkan bahwa GA dan MLR dapat diimplementasikan pada desain obat kanker payudara.

---

ABSTRACTBreast cancer is the first cause of death by cancer in women. Even so, men could have breast cancer. In the treatment of breast cancer there are surgery, radiation therapy and systemic therapy which treatments using drugs. World Health Organization WHO has listed thirty cytotoxic and anticancer drugs to prevent and reduce breast cancer risk. Researchers have been trying to find other drugs to help people with breast cancer. Thus, drug design becomes more important in discovering new potential drugs to treat breast cancer. In this study, we proposed multiple linear regression MLR approach using quantitative structure activity relationship QSAR method for modelling drug design of breast cancer with variable selection using genetic algorithm GA. The obtaining data from public protein bank leads to have lower number of compounds than the number of variables, it failed the assumptions of MLR analysis and led to multicollinearity. QSAR model appeared uncertain when multicollinearity arise. We implemented genetic algorithm GA to resolve multicollinearity. GA acted as a variable selector to obtain the most significant variables and helped getting the most fitted QSAR model. The experimental result shows that combining of GA and MLR can be implemented in breast cancers drug design."

"Penelitian ini bertujuan untuk membuat model peramalan yang efektif dalam meramalkan penjualan produk mobil dalam segmen B2B (Business to Business) agar didapatkan estimasi penjualan produk di masa mendatang. Peneilitian ini menggunakan regresi linear berganda dan jaringan syaraf tiruan yang dioptimasi dengan algoritma genetika.  Faktor peramalan penjualan mobil pada umumnya meliputi penjualan mobil secara nasional, Indeks Harga konsumen, Indeks Kepercayaan Konsumen, Laju Inflasi, Produk Domestik Bruto (GDP), dan  Harga Bahan Bakar Minyak (BBM). Penulis juga telah mendapatkan faktor yang berpengaruh dalam penjualan segmen B2B dengan menyebarkan survey (kuesioner) kepada 102 orang DMU (Decision Making Unit) yang memiliki keputusan dalam pembelanjaan mobil di perusahaan mereka. Kemudian hasil scoring dari kuesioner tersebut kami bobotkan pada data training dan simulasi pada Jaringan Syaraf Tiruan. Hasil penelitian ini menunjukkan bahwa Jaringan Syaraf Tiruan yang dioptimasi  dengan Algoritma Genetika dengan 18 Variabel dapat meningkatkan akurasi peramalan penjualan mobil segmen B2B dengan error 1,3503%, jika dibandingkan nilai error pada Jaringan Syaraf Tiruan biasa sebesar 4,173% dan Regresi Linear Berganda sebesar 17,68%.

---

ABSTRACTThis study aims to create an effective forecasting model in predicting sales of car products in the B2B segment (Business-to-Business) in order to obtain estimates of product sales in the future. This research uses multiple linear regression and artificial neural networks that are optimized by genetic algorithms. Car sales forecasting factors generally include National car sales, Consumer Price Index, Consumer Confidence Index, Inflation Rate, Gross Domestic Product (GDP), and Gasoline Price. The author has also obtained an influential factor in the sale of B2B segments by distributing surveys (questionnaires) to 102 DMU (Decision Making Unit) who have a decision in car purchasing at their company. Then the results of the scoring from the questionnaire are weighted to the training and simulation data on the Artificial Neural Network. The results of this study indicate that the Artificial Neural Network optimized with Genetic Algorithm can improve the accuracy of forecasting B2B segment car sales with an error of 1.3503%, when compared to the error value in the usual Artificial Neural Network of 4.173% and Multiple Linear Regression of 17.68 %."

"Nilai hilang merupakan suatu masalah yang sering dijumpai di berbagai bidang dan harus diatasi untuk memperoleh inferensi statistik yang baik seperti penaksiran parameter. Nilai hilang dapat ditemukan pada setiap jenis data, salah satunya pada jenis data cacah/ count data yang berdistribusi Poisson. Solusi untuk mengatasi masalah nilai hilang berjenis data cacah tersebut dapat diatasi dengan menerapkan teknik imputasi ganda. Teknik imputasi ganda merupakan suatu cara mengatasi nilai hilang dengan mengganti setiap nilai yang hilang dengan beberapa nilai estimasi. Teknik imputasi ganda untuk kasus data cacah terdiri dari tiga tahap utama yaitu tahap imputasi berdasarkan model linier normal, tahap analisis dengan metode generalized linear model Poisson regression dan tahap penggabungan pooling parameter yang didasarkan pada aturan Rubin. Studi ini juga dilengkapi dengan simulasi numerik yang bertujuan untuk komparasi akurasi berdasarkan nilai bias yang dihasilkan. Parameter yang digunakan pada simulasi ini yaitu sebesar 5,10 dan 15 dengan jumlah sampel sebesar 200 untuk tujuan mengaproksimasi sifat kenormalan dan simulasi ini diulang untuk empat skenario yang bertingkat untuk setiap parameter berdasarkan besarnya persentase observasi nilai hilang (0%, 10%, 20% dan 30%). Berdasarkan studi literatur dan simulasi numerik yang dilakukan, solusi yang diajukan untuk mengatasi nilai hilang pada data cacah menghasilkan hasil yang cukup memuaskan terutama saat parameter bernilai besar dan persentase observasi nilai hilang yang kecil. Hal ini diindikasikan dengan ukuran bias dan variansi total dari taksiran rata-rata yang kecil. Namun nilai bias cenderung meningkat seiring meningkatnya persentase observasi nilai yang hilang dan saat nilai parameter yang kecil.

---

Missing values are a problem that is often encountered in various fields and must be addressed to obtain good statistical inference such as parameter estimation. Missing values can be found in any type of data, included count data that has Poisson distributed. One solution to overcome that problem is applying multiple imputation techniques. The multiple imputation technique is a way of dealing with missing values by replacing each missing value with some estimated values. The multiple imputation technique for the case of count data consists of three main stages, namely the imputation stage based on the normal linear model, the analysis stage using the generalized linear model Poisson regression and the last stage is pooling parameter based on Rubins rules. This study is also equipped with numerical simulations which aim to compare accuracy based on the resulting bias value. The parameters used in this simulation are 5, 10 and 15 with a sample size of 200 for the purpose of approximating normal properties and this simulation is repeated for four multilevel scenarios for each parameter based on the percentage of observation of missing values (0%, 10%, 20% and 30%). Based on the study of literature and numerical simulations carried out, the solutions proposed to overcome the missing values in the count data yield satisfactory results, especially when the parameters are large and the percentage of observation of the missing values is small. This is indicated by the size of the bias and the total variance of the small average estimate. But the bias value tends to increase with increasing percentage of observation of missing values and when the parameter values are small."

"This text covers both multiple linear regression and some experimental design models. The text uses the response plot to visualize the model and to detect outliers, does not assume that the error distribution has a known parametric distribution, develops prediction intervals that work when the error distribution is unknown, suggests bootstrap hypothesis tests that may be useful for inference after variable selection, and develops prediction regions and large sample theory for the multivariate linear regression model that has m response variables. A relationship between multivariate prediction regions and confidence regions provides a simple way to bootstrap confidence regions. These confidence regions often provide a practical method for testing hypotheses. There is also a chapter on generalized linear models and generalized additive models. There are many R functions to produce response and residual plots, to simulate prediction intervals and hypothesis tests, to detect outliers, and to choose response transformations for multiple linear regression or experimental design models."

"Kanker merupakan salah satu penyebab kematian yang paling sering terjadi di seluruh dunia. Salah satu jenis kanker yang dapat mengancam terutama pada wanita adalah kanker payudara. Terlambatnya pendeteksian dini pada penderita kanker payudara menyebabkan sulitnya penanganan untuk proses penyembuhan dan besarnya angka kemungkinan kematian. Metode machine learning banyak diaplikasikan dalam kasus pendeteksian dini karena metode machine learning cukup efektif untuk mendiagnosis suatu penyakit. Pada penelitian ini digunakan metode Bayesian Logistic Regression untuk memprediksi kanker payudara. Metode Bayesian digunakan untuk menghitung bobot dari setiap parameter dari data pada regresi logistik. Data yang digunakan pada penelitian ini adalah data Wisconsin Breast Cancer Database (WBCD, 1992) yang dapat diakses melalui UCI Machine Learning Repository. Berdasarkan hasil uji coba, metode Bayesian Logistik Regression memperoleh akurasi sebesar 96,85%, precision, recall dan F-1 score sebsar 95,44%. Hasil simulasi tersebut menunjukkan bahwa Bayesian Logistic Regression cukup baik untuk membantu praktisi medis dalam mendiagnosis kanker payudara.

---

Cancer is one of the most common cause of death in the world. One type of cancer that can be threaten women is breast cancer. The delay in early detection in patient with breast cancer can cause difficulty in recovery process and high mortality rate. Machine learning technique is widely applied in cases of early detection, because machine learning technique is quite effective in diagnose a disease. In this study, the Bayesian Logistic Regression method was used to predict breast cancer. The Bayesian method is used to calculate the weight of each parameter from the data in logistic regression. The data that used in this study is the Wisconsin Breast Cancer Database from UCI Machine Learning Repository. Based on the results of the experiment, Bayesian Logistic Regression method give 96.85% accuracy, and 95,44% precision, recall and F-1 score. These performance results show that the Bayesian Logistic Regression is good enough to help medical experts in diagnosing breast cancer.

"

"ABSTRAKPenelitian ini bertujuan untuk memprediksi tingkat konsumsi gas pipa domestik diIndonesia menggunakan metode Neural Network, ARIMAX, dan Multiple LinearRegression (MLR). Peramalan dilakukan hingga periode Desember 2025 denganmenggunakan data historis tingkat konsumsi gas pipa domestik, inflasi, selisihharga minyak dan gas, serta selisih harga batubara dan gas periode Januari 2007sampai dengan September 2012 sebagai prediktor. Hasilnya metode ARIMAXmemberikan hasil yang paling akurat dengan nilai MAPE 3.89%. Metode NeuralNetwork memberikan hasil forecasting dengan nilai MAPE 6.34%, sedangkanmetode MLR mempunyai tingkat error terbesar dengan MAPE 8.39%. Kapasitasproduksi gas Indonesia cukup besar, tetapi jumlah gas yang dikonsumsi untukkeperluan domestik masih tergolong sedikit. Hasil forecasting ketiga metodemenunjukkan ke depannya tingkat konsumsi gas akan terus meningkat.Perbandingan antara hasil forecasting ketiga metode dan Neraca Gas Indonesiacukup besar. Hal ini menunjukkan meskipun Indonesia memiliki potensicadangan gas alam yang sangat melimpah, tetapi permintaan domestik belumterpenuhi secara maksimal.

---

ABSTRACTThis study aims to predict the level of domestic pipeline gas consumption inIndonesia using Neural Network, ARIMAX, and Multiple Linear Regression(MLR). Forecasting is done until the period of December 2025 using historicaldata of domestic pipeline consumption rate, inflation, the difference price of oiland gas, as well as the difference price of coal and gas from the period January2007 until September 2012 as predictor. The result ARIMAX method gives themost accurate results with the value of MAPE 3.89%. Neural Network methodgives forecasting result with MAPE 6.34%, while the MLR method has the largesterror rate with MAPE 8.39%. Indonesia gas production capacity is quite large, butthe amount of gas consumed for domestic use is still relatively small. The thirdmethod of forecasting results indicate the future gas consumption will continue toincrease. Comparison between the results of the three forecasting methods andNeraca Gas Indonesia is quite large. This shows even though Indonesia has veryabundant potential reserves of natural gas, but domestic demand has not been metmaximally."

"A methodology for characterizing ground water quality of watersheds using hydrochemical data that mingle multiple linear regression and structural equation modeling is presented...."

"Particulate matter is the most prevailing pollutant in Peninsula Malaysia having the highest API value compared to the other criteria pollutants. Long-term exposure to small particles less than 10 micrometres may lead to a marked reduction in life expectancy due to increase cardio-pulmonary and lung cancer mortality. Effective forecasting models at the local level predict the concentrations of particulate matter is crucial as the information generated allows the authority and people within a community to take precautionary measures to avoid exposure to unhealthy levels of air quality and implement strategic measures that improve air quality status. The aim of this study is to establish MLR models for different monsoon seasons with meteorological factors as predictors. Daily observations of PM10 concentrations in Kuala Terengganu, Malaysia from January 2005 to December 2011 were selected for predicting PMl0 concentration level. The MLR models for NEM, Inter Monsoon 1, SWM and Inter Monsoon 2 disclose R2 of 0.68, 0.58, 0.57, and 0.63, respectively. Wind speed, relative humidity and rainfall exhibit negative relationship whilst temperature and atmospheric pressure are directly correlated with PM10 concentrations. In conclusion, the developed MLR models are appropriate for forecasting PM10 concentrations at local level for each monsoon."

"Concrete is a composite building material. Due to its increasing demand in the construction industry, its basic ingredients such as cement, fine aggregate and coarse aggregate have become extremely costly. Studies have been carried out to find better and more economical alternatives to these conventional building materials. One such alternative is fly ash, which can be used to partially replace cement. The main disadvantage of conventional concrete is its brittle failure, which can be avoided by using steel fibers. This study identifies the behavior of concrete with regard to impact resistance and its mechanical properties by adding hooked-end steel fibers at levels of 0, 0.75, 1.15 and 1.55% and partially replacing 40% of the cement with 40% fly ash. In addition to the control concrete, there has been four mixes with respective addition of steel fibers. The behavior of normal and fly ash concrete with steel fibers was compared. The combination of fly ash and steel fibers provided a homogeneous and very rich mix, with a delay in the setting time of the concrete. Of all the mixes, the one containing 40% fly ash and 1.55% steel fibers proved to be the best, with a maximum increase in strength of 17% in compression, 25% in split tension, 30% in flexure and 95% in impact energy at 56 days. A multiple linear regression model was also formulated using SPSS (Statistical Package for Social Sciences) software, through which corresponding equations were developed to predict the strength and energy at 28 and 56 days. The equations were also used to predict the strength of the mixes from other researchers’ experimental work. The predicted results corresponded well with the experimental results and the percentage difference was found to be less than 5%."

"The accuracy of measured traffic flow on a roadway largely depends on the correctness of the PCU factors used for converting traffic counts. PCU is the number of passenger cars that are displaced by a single heavy vehicle of a particular type under prevailing roadway, traffic and control conditions. The aim of the present study is to develop more appropriate models for estimating the equivalency units of different vehicle types on multilane highways, considering the limitations of available methods. Estimation of equivalency units for vehicle types is described by developing speed models based on multiple non-linear regression approaches. The equivalency units estimated by using models are found to be realistic and logical under heterogeneous traffic flow conditions. The PCU values estimated by the multiple non-linear regression method are compared with and found to be relatively higher values than the values obtained by the dynamic PCU. The accuracy of the models is checked by comparing the observed values of speed with estimated speeds. The multiple non-linear regression approach is also used for estimating the equivalency units on six-lane divided highways. Results indicate that the proposed methodology can be used for estimation of equivalency units for vehicle types under mixed traffic conditions."