Hasil Pencarian  ::  Simpan CSV :: Kembali

"This book presents the fundamentals of rule learning as investigated in classical machine learning and modern data mining. It introduces a feature-based view, as a unifying framework for propositional and relational rule learning, thus bridging the gap between attribute-value learning and inductive logic programming, and providing complete coverage of most important elements of rule learning."

"Menurut EEB Laboratory Jakarta, pada tahun 2016 sektor bangunan memiliki mengkonsumsi 18-20% dari total penggunaan energi di Indonesia, dan terus menerus meningkat seiring perkembangan teknologi yang membutuhkan sumber energi dalam upaya peningkatan kualitas hidup penghuni bangunan. Bangunan pintar merupakan sebuah konsep pemanfaatan teknologi yang tidak hanya bertujuan meningkatkan kenyamanan penghuni, tetapi juga dapat membantu dalam upaya efisiensi energi pada operasional bangunan. Maka dari itu, penelitian ini akan membantu upaya perancangan efisiensi energi pada sebuah bangunan dengan meninjau fitur dan karakteristik yang berpotensi dalam mendukung efisiensi energi dengan penerapan konsep bangunan pintar. Selain itu, akan dibuat sebuah model dengan pemanfaatan machine learning yang mampu memberikan prediksi tingkat penggunaan energi berdasarkan fitur-fitur yang diberikan. Model machine learning yang dihasilkan memiliki rata-rata nilai kesalahan relatif sebesar 17,76%, serta didapatkan tingkat efisiensi dengan penerapan seluruh fitur yang diidentifikasi pada rentang 34,5% hingga 45,3% tergantung pada lantai yang ditinjau.

---

According to EEB Laboratory Jakarta, Indonesian building sector accounts for 18- 20% energy consumption in 2016, and this trend will continuously increase as technology needed to increase housing residents' quality keeps advancing. Smart building is a concept to utilise technology that does not only help increase occupants' comfort inside the building, but it can also help increase energy usage efficiency in building operations. This research aims to help the effort in designing energy efficiency planning for a building by reviewing potential features and characteristics that could help improves energy efficiency with implementation of the smart building concept. A model based on machine learning that could give prediction on the level of energy consumption based on given features will also be discussed here. This model of machine learning has a 17,76% average of relative error, as well as 34,5% until 45,3% efficienct level that includes implementation of all features, depending on analysed floor."

"This book describes efforts to improve subject-independent automated classification techniques using a better feature extraction method and a more efficient model of classification. It evaluates three popular saliency criteria for feature selection, showing that they share common limitations, including time-consuming and subjective manual de-facto standard practice, and that existing automated efforts have been predominantly used for subject dependent setting. It then proposes a novel approach for anomaly detection, demonstrating its effectiveness and accuracy for automated classification of biomedical data, and arguing its applicability to a wider range of unsupervised machine learning applications in subject-independent settings."

"Meskipun kajian mengenai bahan bakar dan penggerak alternatif sudah banyak, namun target dan aplikasinya dalam pengurangan emisi CO2 di pelabuhan masih kurang mendapat perhatian terutama di negara berkembang seperti Indonesia. Penelitian ini menggunakan machine learning dalam memperkirakan emisi CO2 dari aktivitas kapal di tujuh pelabuhan di Indonesia kemudian dicari variable yang berpengaruh pada total emisi sebagai fokus dalam pengembangan Ship Energy Efficiency Management Plan (SEEMP). Dilakukan prediksi total emisi CO2 menggunakan regresi hutan acak kemudian keefektifannya diverifikasi menggunakan validasi silang k-fold, hasil prediksi kemudian dibandingkan dengan total emisi perhitungan metode bottom-up. Hasil analisis attribute weight berdasarkan correlation menunjukkan bahwa daya mesin dan waktu operasi kapal di pelabuhan memiliki pengaruh yang lebih besar dalam menghasilkan emisi CO2. Prediksi total emisi menunjukkan bahwa model memiliki akurasi yang cukup rendah akibat banyaknya data yang kosong meskipun algoritma model sudah tergolong bagus. Akhirnya, operasi hemat bahan bakar dibahas dengan fokus pada tenaga dan bahan bakar alternatif serta peningkatan efisiensi kerja, penggunaan bahan bakar bersih dari hidrogen dan biofuel mamiliki potensi pengurangan yang paling tinggi dengan cold ironing sebagai alternatif yang dapat memenuhi syarat pengurangan emisi per tahun sebesar 20%. Dibutuhkan data yang lengkap untuk melakukan prediksi total emisi yang akurat serta pengembangan teknis dan ketersediaan sumber daya pada metode yang telah dibahas agar dapat di implementasikan kedalam Rencana Pengelolaan Efisiensi Energi Kapal.

---

Although there are many studies on alternative fuels and drivers, the target and their application in reducing CO2 emissions at ports have received little attention, especially in developing countries such as Indonesia. This study uses machine learning to estimate CO2 emissions from ship activities at seven ports in Indonesia and then looks for variables that affect total emissions as a focus in developing a Ship Energy Efficiency Management Plan (SEEMP). Total CO2 emissions were predicted using random forest regression, their effectiveness was then verified using k-fold cross-validation, the prediction results were then compared with the total emissions calculated using the bottom-up method. The results of attribute weight analysis based on correlation show that engine power and ship operating time in port have a greater influence in producing CO2 emissions. Prediction of total emissions shows that the model has a fairly low accuracy due to the large number of blank data despite the model algorithm exelency. Finally, fuel-efficient operations are discussed with a focus on alternative power and fuels as well as improving work efficiency, the use of clean fuels from hydrogen and biofuels has the highest reduction potential with cold ironing as an alternative that can meet the requirements of 20% annual emission reduction. Complete data is needed to make accurate predictions of total emissions as well as technical development and resource availability on the methods discussed so that they can be implemented into the Ship Energy Efficiency Management Plan."

"Analisis sentimen merupakan permasalahan klasifikasi data mining dengan proses memahami, mengekstrak dan mengolah data teks secara otomatis untuk mendapatkan informasi. Dalam menganalisis pendapat di media sosial digunakan machine learning untuk mendapatkan hasil klasifikasi. Banyak metode machine learning untuk melakukan klasifikasi, dalam penelitian ini akan digunakan convolutional neural network. Dalam machine learning, data dibagi menjadi data training dan data test dengan domain data yang sama. Permasalahan utama skripsi ini adalah data yang digunakan memiliki dua domain berbeda, sehingga metode machine learning tradisional tidak dapat diterapkan. Sehingga agar dapat menerapkan convolutional neural network untuk dua data berbeda diperkenalkan suatu cara yaitu transfer learning. Transfer learning merupakan suatu proses pembelajaran model yang didapatkan dari training data A oleh data B dengan domain berbeda. Simulasi dalam penelitian ini menghasilkan suatu akurasi transfer learning dengan metode convolutional neural network.

---

Sentiment analysis is classification problem in data mining with process of understanding, extracting and processing text data to obtain information. Machine learning is needed in analyzing sentiment of the people to get the result of classification. There are many methods in machine learning to do classification, this research will use convolutional neural network. In machine learning, data is divided into train and test data with the same domain.

The main problem of this research is the data has a different domain, so the traditional machine learning method can not be applied. In order to apply convolutional neural network into data with different domain, it will be introduced transfer learning method. Transfer learning is learning model process obtained from training data A then tested by data B. In this research, the simulations result is accuracy of transfer learning with convolutional neural network."

"This two-volume set LNAI 7523 and LNAI 7524 constitutes the refereed proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases: ECML PKDD 2012, held in Bristol, UK, in September 2012. The 105 revised research papers presented together with 5 invited talks were carefully reviewed and selected from 443 submissions. The final sections of the proceedings are devoted to Demo and Nectar papers. The Demo track includes 10 papers (from 19 submissions) and the Nectar track includes 4 papers (from 14 submissions). The papers grouped in topical sections on association rules and frequent patterns, Bayesian learning and graphical models, classification, dimensionality reduction, feature selection and extraction, distance-based methods and kernels, ensemble methods, graph and tree mining, large-scale, distributed and parallel mining and learning, multi-relational mining and learning, multi-task learning, natural language processing, online learning and data streams, privacy and security, rankings and recommendations, reinforcement learning and planning, rule mining and subgroup discovery, semi-supervised and transductive learning, sensor data; sequence and string mining, social network mining, spatial and geographical data mining, statistical methods and evaluation, time series and temporal data mining, and transfer learning."

"This two-volume set LNAI 7523 and LNAI 7524 constitutes the refereed proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases: ECML PKDD 2012, held in Bristol, UK, in September 2012. The 105 revised research papers presented together with 5 invited talks were carefully reviewed and selected from 443 submissions. The final sections of the proceedings are devoted to Demo and Nectar papers. The Demo track includes 10 papers (from 19 submissions) and the Nectar track includes 4 papers (from 14 submissions). The papers grouped in topical sections on association rules and frequent patterns, Bayesian learning and graphical models, classification, dimensionality reduction, feature selection and extraction; distance-based methods and kernels, ensemble methods; graph and tree mining, large-scale, distributed and parallel mining and learning, multi-relational mining and learning, multi-task learning, natural language processing, online learning and data streams, privacy and security, rankings and recommendations, reinforcement learning and planning, rule mining and subgroup discovery; semi-supervised and transductive learning; sensor data, sequence and string mining, social network mining, spatial and geographical data mining, statistical methods and evaluation, time series and temporal data mining, and transfer learning."

"This volume offers comprehensive coverage of state-of-the-art ensemble learning techniques, including the random forest skeleton tracking algorithm in the Xbox Kinect sensor, which bypasses the need for game controllers. At once a solid theoretical study and a practical guide, the volume is a windfall for researchers and practitioners alike. "

"This book constitutes the refereed proceedings of the 8th International Conference, MLDM 2012, held in Berlin, Germany in July 2012. The 51 revised full papers presented were carefully reviewed and selected from 212 submissions. The topics range from theoretical topics for classification, clustering, association rule and pattern mining to specific data mining methods for the different multimedia data types such as image mining, text mining, video mining and web mining."

"Discover the practical aspects of implementing deep-learning solutions using the rich Python ecosystem. This book bridges the gap between the academic state-of-the-art and the industry state-of-the-practice by introducing you to deep learning frameworks such as Keras, Theano, and Caffe. The practicalities of these frameworks is often acquired by practitioners by reading source code, manuals, and posting questions on community forums, which tends to be a slow and a painful process.Deep Learning with Python allows you to ramp up to such practical know-how in a short period of time and focus more on the domain, models, and algorithms. This book briefly covers the mathematical prerequisites and fundamentals of deep learning, making this book a good starting point for software developers who want to get started in deep learning. A brief survey of deep learning architectures is also included. Deep Learning with Python also introduces you to key concepts of automatic differentiation and GPU computation which, while not central to deep learning, are critical when it comes to conducting large scale experiments. You will: Leverage deep learning frameworks in Python namely, Keras, Theano, and Caffe Gain the fundamentals of deep learning with mathematical prerequisites Discover the practical considerations of large scale experiments Take deep learning models to production"