Concept Learning

Shultz, Thomas R.; Fahlman, Scott E.; Craw, Susan; Andritsos, Periklis; Tsaparas, Panayiotis; Silva, Ricardo; Drummond, Chris; Lanzi, Pier Luca; Gama, João; Wiegand, R. Paul; Sen, Prithviraj; Namata, Galileo; Bilgic, Mustafa; Getoor, Lise; He, Jiang; Stephan, Frank; Harries, Michael; Ting, Kai Ming; Pfahringer, Bernhard; Case, John; Jain, Sanjay; Wagstaff, Kiri L.; Nijssen, Siegfried; Wirth, Anthony; Ling, Charles X.; Sheng, Victor S.; Zhang, Xinhua; Sammut, Claude; Cancedda, Nicola; Renders, Jean-Michel; Michelucci, Pietro; Oblinger, Daniel; Keogh, Eamonn; Mueen, Abdullah

doi:10.1007/978-0-387-30164-8_154

Cited by 31 publications

(35 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FN is the event that the model predicts the sample does not belong to a label when the sample has the label. Positive predictive value (PPV) is therefore defined as (Sammut & Webb, 2011):

{\rm{PPV}} = \frac{{{\rm{TP}}}}{{{\rm{TP}} + {\rm{FP}}}} = \frac{{{\rm{TP}}}}{{{\rm{all}}\,{\rm{positive}}\,{\rm{responses}}}}

. Moreover, the complement of the PPV, the false discovery rate (FDR) is therefore defined as

{\rm{FDR}} = {\rm{1}} - {\rm{PPV}} = \frac{{{\rm{FP}}}}{{{\rm{TP}} + {\rm{FP}}}} = \frac{{{\rm{FP}}}}{{{\rm{all}}\,{\rm{positive}}\,{\rm{responses}}}}

.…”

Section: Methodsmentioning

confidence: 99%

Inferring geological structural features from geophysical and geological mapping data using machine learning algorithms

Green

2023

Geophysical Prospecting

View full text Add to dashboard Cite

We present an automated approach for inferring surface geological structures from geophysical survey data. Our method employs machine learning, using mapped geological structures as labels and filtered geophysical surveys as reference maps. We compared the performance of the eight main machine learning algorithms and their 32 branches. Applied to the Geological Survey of Victoria's database for the Bendigo Zone, following an appropriate choice of geological features, the 3‐class classification model using subspace K‐nearest neighbour methods achieves a stable and validated 92% accuracy in around 1 min. The fault‐only classification model achieves a stable and validated 97% accuracy in around 6 min. This shows that geological structural features on the surface may be inferred from between one and three of the following geophysical data types: gravity, airborne total magnetic intensity and first vertical derivative of total magnetic intensity. It shows the prospect of machine learning in geological research and suggests that geophysical data combined with machine learning may be useful and efficient in determining the existence of geological structural features.

show abstract

“…FN is the event that the model predicts the sample does not belong to a label when the sample has the label. Positive predictive value (PPV) is therefore defined as (Sammut & Webb, 2011):

{\rm{PPV}} = \frac{{{\rm{TP}}}}{{{\rm{TP}} + {\rm{FP}}}} = \frac{{{\rm{TP}}}}{{{\rm{all}}\,{\rm{positive}}\,{\rm{responses}}}}

. Moreover, the complement of the PPV, the false discovery rate (FDR) is therefore defined as

{\rm{FDR}} = {\rm{1}} - {\rm{PPV}} = \frac{{{\rm{FP}}}}{{{\rm{TP}} + {\rm{FP}}}} = \frac{{{\rm{FP}}}}{{{\rm{all}}\,{\rm{positive}}\,{\rm{responses}}}}

.…”

Section: Methodsmentioning

confidence: 99%

Inferring geological structural features from geophysical and geological mapping data using machine learning algorithms

Green

2023

Geophysical Prospecting

View full text Add to dashboard Cite

show abstract

“…ML methods can have more meaningful results for various research topics [3,[22][23][24]. The important thing is to determine the correct algorithms to obtain the relevant results for the problem [9,[25][26][27]. ANN and fuzzy inference systems were preferred for this study because of their ability to learn and predict complex and nonlinear relationships.…”

Section: Machine Learning Algorithmsmentioning

confidence: 99%

Estimation of UAV flight time and Battery Consumption for photogrammetric application using multiple machine learning algorithms

2022

View full text Add to dashboard Cite

In recent years, important research has been conducted in Machine Learning (ML), especially on Artificial Neural Networks (ANN). Adaptive-Network Based Fuzzy Inference Systems (ANFIS) and Particle Swarm Optimization-Fuzzy Inference System (PSO-FIS) algorithms are popular ML algorithms like ANN. In terms of their working architecture and results, ANN, ANFIS, and PSO-FIS algorithms can obtain useful solutions for different nonlinear problems. This study evaluated the performance of the ANN, ANFIS, and PSO-FIS algorithms and compared the estimation results. Regarding the application, the test and target data was obtained from the flights performed with Unmanned Aerial Vehicles (UAV), including how long the UAV operates (i.e., Flight Time, FT) and how much battery the UAV consumes during the flight (i.e., Battery Consumption, BC). To obtain FT and BC outputs, sixty-five pre- and post-flight data tables were created. The best iterations for estimating the outputs using the three ML algorithms (considering the minimum/maximum values, RMSE, R, and R2) were determined and discussed based on the training, validation, and test estimations.

show abstract

“…Support Vector Machine (SVM): This is thought to be the optimal classifier for determining binary outcomes, such as benignity and malignancy. The theoretical ''hyperplane'' that separates these 2 outcomes exists in a multi-dimensional space, which consists of as many dimensions as there are the number of parameters [22,23]. 4.…”

Section: Decision Tree (Dt)mentioning

confidence: 99%

“…1 Flow chart of inclusion criteria for training and testing groups dimensional spaces, with each parameter considered to add one dimension, was introduced. To train and internally validate our predictive model as well as overcome dataset biases, a resampling technique known as k fold cross-validation was employed [23]. This technique randomly partitions the training group into k fold subsamples (k =1 0i n this case).…”

Section: Statistical Analysis and Artificial Intelligence Modelmentioning

confidence: 99%

“…Several measures of predictive performance were calculated, including the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity, and the F-score. The F-score is a measure of accuracy in binary classification, including both precision and recall [23]. Where numbers were too low to populate the confusion matrix for sub-group analysis, the percentage of correctly classified cases was reported instead.…”

Section: Statistical Analysis and Artificial Intelligence Modelmentioning

confidence: 99%

See 1 more Smart Citation

Artificial Intelligence for Pre‐operative Diagnosis of Malignant Thyroid Nodules Based on Sonographic Features and Cytology Category

et al. 2022

View full text Add to dashboard Cite

Background Current diagnosis and classification of thyroid nodules are susceptible to subjective factors. Despite widespread use of ultrasonography (USG) and fine needle aspiration cytology (FNAC) to assess thyroid nodules, the interpretation of results is nuanced and requires specialist endocrine surgery input. Using readily available pre-operative data, the aims of this study were to develop artificial intelligence (AI) models to classify nodules into likely benign or malignant and to compare the diagnostic performance of the models. Methods Patients undergoing surgery for thyroid nodules between 2010 and 2020 were recruited from our institution’s database into training and testing groups. Demographics, serum TSH level, cytology, ultrasonography features and histopathology data were extracted. The training group USG images were re-reviewed by a study radiologist experienced in thyroid USG, who reported the relevant features and supplemented with data extracted from existing reports to reduce sampling bias. Testing group USG features were extracted solely from existing reports to reflect real-life practice of a non-thyroid specialist. We developed four AI models based on classification algorithms (k-Nearest Neighbour, Support Vector Machine, Decision Tree, Naïve Bayes) and evaluated their diagnostic performance of thyroid malignancy. Results In the training group (n = 857), 75% were female and 27% of cases were malignant. The testing group (n = 198) consisted of 77% females and 17% malignant cases. Mean age was 54.7 ± 16.2 years for the training group and 50.1 ± 17.4 years for the testing group. Following validation with the testing group, support vector machine classifier was found to perform best in predicting final histopathology with an accuracy of 89%, sensitivity 89%, specificity 83%, F-score 94% and AUROC 0.86. Conclusion We have developed a first of its kind, pilot AI model that can accurately predict malignancy in thyroid nodules using USG features, FNAC, demographics and serum TSH. There is potential for a model like this to be used as a decision support tool in under-resourced areas as well as by non-thyroid specialists.

show abstract

Concept Learning

Cited by 31 publications

References 6 publications

Inferring geological structural features from geophysical and geological mapping data using machine learning algorithms

Inferring geological structural features from geophysical and geological mapping data using machine learning algorithms

Estimation of UAV flight time and Battery Consumption for photogrammetric application using multiple machine learning algorithms

Artificial Intelligence for Pre‐operative Diagnosis of Malignant Thyroid Nodules Based on Sonographic Features and Cytology Category

Contact Info

Product

Resources

About