Unintended Consequences of Machine Learning in Medicine Cabitza, Federico; Rasoini, Raffaele; Gensini, Gian Franco
JAMA : the journal of the American Medical Association,
08/2017, Letnik:
318, Številka:
6
Journal Article
Recenzirano
The article discusses the growing interest in machine learning seen in the field of medicine as is in many other sectors as well. Some of the unintended consequences of machine learning in medicine ...are highlighted.
In a previous study conducted by (Cut Nuraini, "Combine Application Theory of Humanistic Learning with Cognitive Methods in Learning (Its Application in Learning Indonesian Language) as One Model of ...Teaching Arts)". UNSIKA: 2016) have obtained quite good results in the application of Combine Application of Humanistic and Cognitive theories in Indonesian Language Learning at FASILKOM. Referring to the success of previous research, it is unfortunate if it is not continued and developed. Previous research development in this study was by designing book applications in digital form or applications from the application of Cognitive Theory and Humastik on Indonesian Language learning at FASILKOM using the Forward Chaining method in Expert Systems.
The task of algorithm selection involves choosing an algorithm from a set of algorithms on a per-instance basis in order to exploit the varying performance of algorithms over a set of instances. The ...algorithm selection problem is attracting increasing attention from researchers and practitioners in AI. Years of fruitful applications in a number of domains have resulted in a large amount of data, but the community lacks a standard format or repository for this data. This situation makes it difficult to share and compare different approaches effectively, as is done in other, more established fields. It also unnecessarily hinders new researchers who want to work in this area. To address this problem, we introduce a standardized format for representing algorithm selection scenarios and a repository that contains a growing number of data sets from the literature. Our format has been designed to be able to express a wide variety of different scenarios. To demonstrate the breadth and power of our platform, we describe a study that builds and evaluates algorithm selection models through a common interface. The results display the potential of algorithm selection to achieve significant performance improvements across a broad range of problems and algorithms.
Augmented reality (AR) has proven to be an invaluable interactive medium to reduce cognitive load by bridging the gap between the task-at-hand and relevant information by displaying information ...without disturbing the user's focus. AR is particularly useful in the manufacturing environment where a diverse set of tasks such as assembly and maintenance must be performed in the most cost-effective and efficient manner possible. While AR systems have seen immense research innovation in recent years, the current strategies utilised in AR for camera calibration, detection, tracking, camera position and orientation (pose) estimation, inverse rendering, procedure storage, virtual object creation, registration, and rendering are still mostly dominated by traditional non-AI approaches. This restricts their practicability to controlled environments with limited variations in the scene. Classical AR methods can be greatly improved through the incorporation of various AI strategies like deep learning, ontology, and expert systems for adapting to broader scene variations and user preferences. This research work provides a review of current AR strategies, critical appraisal for these strategies, and potential AI solutions for every component of the computational pipeline of AR systems. Given the review of current work in both fields, future research work directions are also outlined.
•Comparison of sequence learner and static learner on real-world dataset.•Sequence learner improves fraud detection on offline transactions.•Both sequence learner and static learner benefit from ...manual feature aggregations.•Frauds detected by sequence learner and static learner are consistently different.
Due to the growing volume of electronic payments, the monetary strain of credit-card fraud is turning into a substantial challenge for financial institutions and service providers, thus forcing them to continuously improve their fraud detection systems. However, modern data-driven and learning-based methods, despite their popularity in other domains, only slowly find their way into business applications.
In this paper, we phrase the fraud detection problem as a sequence classification task and employ Long Short-Term Memory (LSTM) networks to incorporate transaction sequences. We also integrate state-of-the-art feature aggregation strategies and report our results by means of traditional retrieval metrics.
A comparison to a baseline random forest (RF) classifier showed that the LSTM improves detection accuracy on offline transactions where the card-holder is physically present at a merchant. Both the sequential and non-sequential learning approaches benefit strongly from manual feature aggregation strategies. A subsequent analysis of true positives revealed that both approaches tend to detect different frauds, which suggests a combination of the two. We conclude our study with a discussion on both practical and scientific challenges that remain unsolved.
Accurately forecasting electricity prices is essential for a variety of stakeholders in the energy sector, including market investors, policymakers, and consumers. However, existing forecasting ...techniques are often limited by complex parametric estimates and strict restrictions on input variables. This paper proposes a Whale Optimization Algorithm (WOA)-based multivariate exponential smoothing Grey-Holt (GMHES) model for electricity price forecasting. The proposed WOA-GMHES(1,N) model uses historical data to learn the underlying trends and patterns of electricity prices. The WOA algorithm is used to optimize the model parameters, which are adaptively adjusted to reflect the changing dynamics of the electricity market. The proposed model is evaluated on real high- and low-voltage electricity price data from Cameroon. The results show that the novel WOA-GMHES(1,N) model outperforms competing models, achieving RMSE and SMAPE scores of 0.1359 and 0.61%, respectively. This novel model is also computationally efficient, requiring less than 1.3 s to generate a forecast. The proposed WOA-GMHES(1,N) model is a promising novel approach for electricity price forecasting. The model is accurate, efficient, and flexible, making it a valuable tool for a variety of stakeholders in the energy sector.
•We do image classification on training data limited dataset with deep learning.•Transfer learning is employed to overcome the serious over-fitting.•Web data augmentation is developed to improve the ...classification performance.•Bayesian optimization is employed to facilitate the hyper-parameter search.
Since Convolutional Neural Network (CNN) won the image classification competition 202 (ILSVRC12), a lot of attention has been paid to deep layer CNN study. The success of CNN is attributed to its superior multi-scale high-level image representations as opposed to hand-engineering low-level features. However, estimating millions of parameters of a deep CNN requires a large number of annotated samples, which currently prevents many superior deep CNNs (such as AlexNet, VGG, ResNet) being applied to problems with limited training data. To address this problem, a novel two-phase method combining CNN transfer learning and web data augmentation is proposed. With our method, the useful feature presentation of pre-trained network can be efficiently transferred to target task, and the original dataset can be augmented with the most valuable Internet images for classification. Our method not only greatly reduces the requirement of a large training data, but also effectively expand the training dataset. Both of method features contribute to the considerable over-fitting reduction of deep CNNs on small dataset. In addition, we successfully apply Bayesian optimization to solve the tuff problem, hyper-parameter tuning, in network fine-tuning. Our solution is applied to six public small datasets. Extensive experiments show that, comparing to traditional methods, our solution can assist the popular deep CNNs to achieve better performance. Particularly, ResNet can outperform all the state-of-the-art models on six small datasets. The experiment results prove that the proposed solution will be the great tool for dealing with practice problems which are related to use deep CNNs on small dataset.
Every year, millions of people worldwide (including a major portion in China) are suffering from lung cancer disease (Chinese report of Smoking and Health 2017). The aim of this paper is to develop a ...new fuzzy soft expert system which can be used to predict lung cancer disease. A prediction process using this fuzzy soft expert system is composed of four main steps: (1) Transform real-valued inputs into fuzzy numbers. (2) Transform fuzzy numbers of data into fuzzy soft sets. (3) Reduce, using normal parameter reduction method, the obtained family of fuzzy soft sets into a new family of fuzzy soft sets. (4) Use the proposed algorithm to get the output data. An experiment is conducted on forty five patients (thirty males, fifteen females, all are cigarette smokers) who endure treatment in the Respiratory Department of Nanjing Chest Hospital, China. The number of training data taken was 55 records, and the remaining 45 records were used for the testing process in our system by using weight loss, shortness of breath, chest pain, persistence a cough, blood in sputum, and age of patients. The quantized accuracies of the proposed system were found to be
100
%
. In this work, we developed a fuzzy soft expert system based on fuzzy soft sets; we used a fuzzy membership functions and an algorithm to predict those patients who may suffer lung cancer. In this way, it is possible to conclude that the use of fuzzy soft expert system can produce valuable results for lung cancer detection. It is found that the fuzzy soft expert system developed is useful to the expert doctor to decide if a patient has lung cancer or not. Finally, we introduce comparison diagnosed between our proposed system and the fuzzy inference system.
Abstract
The article contains the digitalization results for expert systems design in the framework of geoinformation management within Arctic while coronavirus and change of climate. Currently, the ...methods of designing expert systems within the framework of geoinformation management in the Arctic demonstrate the use of improved concepts of data collection and visualization. The article gives preference to the use of web constructors in distributed networks. An example of a working layout of an expert system for geoinformation support of the activities of a subarctic port in the conditions of ice waters is given. The results of the research can be useful in operational environmental activity and for university learning purposes, including Master’s programs.
Decision trees: a recent overview Kotsiantis, S. B.
The Artificial intelligence review,
04/2013, Letnik:
39, Številka:
4
Journal Article
Recenzirano
Decision tree techniques have been widely used to build classification models as such models closely resemble human reasoning and are easy to understand. This paper describes basic decision tree ...issues and current research points. Of course, a single article cannot be a complete review of all algorithms (also known induction classification trees), yet we hope that the references cited will cover the major theoretical issues, guiding the researcher in interesting research directions and suggesting possible bias combinations that have yet to be explored.