@phdthesis{oai:kyutech.repo.nii.ac.jp:00006003,
 author = {Yelameli,  Mallikarjun Rajendrakumar},
 month = {2019-06-13},
 note = {1 Introduction||2 Machine Learning Models for Classification||3 Result and Discussion||4 Conclusion and Future Scope, The aim is to study the use of machine learning algorithms for the classification hydrothermal seafloor rocks measured underwater using Laser-Induced Breakdown Spectroscopy.  The rocks were classified concerning their labels assigned to each rock, and geological groups formed ternary diagram with the relative ratio of Cu-Pb-Zn. In this research the target rocks are obtained from deep-ocean in Okinawa Japan. These were hydrothermal deposit sea-floor rocks.  Further, these rocks were brought into the laboratory and broken into pieces and made the pellets. The experimental setup which resembles the ocean, then to test using ChemiCam device which is a LIBS device which is specifically designed for the chemical elemental analysis in the deep ocean is used to fire the laser beams on rocks. The proposed methods for classification of rocks with respect to their labels and for geological group are evaluated using with and without linear detrend along with the principal component analysis （PCA） as a pre-processing step which significantly reduces the dimensionality of the data, with classification algorithms such as the support vector machine （SVM）, k-nearest neighbor search （KNN） and artificial neural network （ANN） methods. The performance of the classification algorithms depends on the size of the dataset, to prove this the dataset has been divided into two sets of 100 laser shots of each rock and 300 laser shots of each rock.  Additionally, removing the noise from the spectra such as linear trend using linear detrend operation from the data enhances the performance of the classification in terms of sensitivity. The best classification performance concerning the rock label concerning sensitivity is obtained using an SVM linear kernel algorithm with 95%. The best classification performance concerning the geological group is obtained using the SVM method with 98% accuracy. The one-sided Wilcoxon signed rank test is applied to the classification results in the rock label and group cases, and the results indicate that the SVM algorithm has statistical significance over the other algorithms while classifying the rock labels and rock group., 九州工業大学博士学位論文 学位記番号：生工博甲第345号 学位授与年月日：平成31年3月25日, 平成30年度},
 school = {九州工業大学},
 title = {Study on Machine Learning Algorithms and Statistical Analysis for Classification of Hydrothermal Seafloor Rocks Measured Underwater Using Laser-Induced Breakdown Spectroscopy},
 year = {}
}