Feature shaping for linear SVM classifiers

2009 ◽  
Author(s):  
George Forman ◽  
Martin Scholz ◽  
Shyamsundar Rajaram
Keyword(s):  
Linear Svm

Visible/Near infrared (VIS/NIR) spectroscopy and multivariate data analysis (MVDA) for identification and quantification of olive leaf spot (OLS) disease

2014 ◽  
Vol 2 (1) ◽  
pp. 1-9
Author(s):  
Nawaf Abu-Khalaf ◽  
Mazen Salman
Keyword(s):  
Leaf Spot ◽  
Near Infrared ◽  
Latent Infection ◽  
Linear Method ◽  
Nir Spectroscopy ◽  
Support Vector ◽  
Olive Leaf ◽  
Classification Rate ◽  
Infection Period ◽  
Linear Svm

Early detection of plant disease requires usually elaborating methods techniques and especially when symptoms are not visible. Olive Leaf Spot (OLS) infecting upper surface of olive leaves has a long latent infection period. In this work, VIS/NIR spectroscopy was used to determine the latent infection and severity of the pathogens. Two different classification methods were used, Partial Least Squared-Discrimination Analysis (PLS-DA) (linear method) and Support Vector Machine (SVM) (non-linear). SVM-classification was able to classify severity levels 0, 1, 2, 3, 4, and 5 with classification rates of 94, 90, 73, 79, 83 and 100%, respectively The overall classification rate was about 86%. PLS-DA was able to classify two different severity groups (first group with severity 0, 1, 2, 3, and second group with severity 4, 5), with a classification rate greater than 95%. The results promote further researches, and the possibility of evaluation OLS in-situ using portable VIS/NIR devices.

Visible/Near infrared (VIS/NIR) spectroscopy and multivariate data analysis (MVDA) for identification and quantification of olive leaf spot (OLS) disease

2014 ◽  
Vol 2 (1) ◽  
pp. 1-9
Author(s):  
Nawaf Abu-Khalaf ◽  
Mazen Salman
Keyword(s):  
Leaf Spot ◽  
Near Infrared ◽  
Latent Infection ◽  
Linear Method ◽  
Nir Spectroscopy ◽  
Support Vector ◽  
Olive Leaf ◽  
Classification Rate ◽  
Infection Period ◽  
Linear Svm

Early detection of plant disease requires usually elaborating methods techniques and especially when symptoms are not visible. Olive Leaf Spot (OLS) infecting upper surface of olive leaves has a long latent infection period. In this work, VIS/NIR spectroscopy was used to determine the latent infection and severity of the pathogens. Two different classification methods were used, Partial Least Squared-Discrimination Analysis (PLS-DA) (linear method) and Support Vector Machine (SVM) (non-linear). SVM-classification was able to classify severity levels 0, 1, 2, 3, 4, and 5 with classification rates of 94, 90, 73, 79, 83 and 100%, respectively The overall classification rate was about 86%. PLS-DA was able to classify two different severity groups (first group with severity 0, 1, 2, 3, and second group with severity 4, 5), with a classification rate greater than 95%. The results promote further researches, and the possibility of evaluation OLS in-situ using portable VIS/NIR devices.

scholarly journals Data-mining Techniques for Image-based Plant Phenotypic Traits Identification and Classification

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Md. Matiur Rahaman ◽  
Md. Asif Ahsan ◽  
Ming Chen
Keyword(s):  
Data Mining ◽  
Agricultural Development ◽  
Computational Time ◽  
Support Vector ◽  
Phenotypic Traits ◽  
Linear Discriminant ◽  
Plant Phenomics ◽  
Classification Prediction ◽  
Linear Svm ◽  
Integrated Or

AbstractStatistical data-mining (DM) and machine learning (ML) are promising tools to assist in the analysis of complex dataset. In recent decades, in the precision of agricultural development, plant phenomics study is crucial for high-throughput phenotyping of local crop cultivars. Therefore, integrated or a new analytical approach is needed to deal with these phenomics data. We proposed a statistical framework for the analysis of phenomics data by integrating DM and ML methods. The most popular supervised ML methods; Linear Discriminant Analysis (LDA), Random Forest (RF), Support Vector Machine with linear (SVM-l) and radial basis (SVM-r) kernel are used for classification/prediction plant status (stress/non-stress) to validate our proposed approach. Several simulated and real plant phenotype datasets were analyzed. The results described the significant contribution of the features (selected by our proposed approach) throughout the analysis. In this study, we showed that the proposed approach removed phenotype data analysis complexity, reduced computational time of ML algorithms, and increased prediction accuracy.

scholarly journals SVM ALGORITHM FOR PREDICTING RICE YIELDS

2020 ◽  
Vol 13 (2) ◽  
pp. 50-54
Author(s):  
Nur Nafi'iyah
Keyword(s):  
Agricultural Land ◽  
Rice Yield ◽  
Weather Conditions ◽  
Original Data ◽  
Research Process ◽  
Svm Algorithm ◽  
Food Shortages ◽  
Rice Yields ◽  
The Government ◽  
Linear Svm

Agriculture in Indonesia is highly dependent on reservoir irrigation water sources and rain. Because some agricultural land in Indonesia is rainfed. Plants in Indonesia rely on water from rain and irrigation. Weather conditions greatly affect the number of farmers' harvest. Farmers often experience crop failures due to changing weather. From data from the Central Statistics Agency, it is stated that the number of rice yields in 2019 decreased by 7.76% compared to 2018. In order to avoid rice imports and rice food shortages, a breakthrough is needed that can help the government in making policies. One of the breakthroughs is creating a rice yield prediction system. The research process consisted of collecting data via the web: https://www.pertanian.go.id/home/?show=page&act=view&id=61. The data shows the variables of province, year, land area, production. The total number of data is 170 rows, with a division of 130 lines for training, and 40 for testing. Furthermore, the data is processed and processed and normalized. The results of data processing are then trained and predicted with a linear SVM kernel. The results of SVM prediction with original data without normalization of MAPE 6635.53%, and RMSE 1094810.74. The results of SVM prediction with normalized data first, the MAPE value was 9427.714%, and RMSE 0.017.

Sign in / Sign up

Export Citation Format

Share Document