•A financial expert system for predicting the monthly stock price.•“Knowledge base” captures the impact of traditional and online sources.•The “engine” uses four ensemble machine learning ...methods.•The error for most ensembles considered is < 1%.•The system is hosted online and freely available.
With the ubiquity of the Internet, platforms such as: Google, Wikipedia and the like can provide insights pertaining to firms’ financial performance as well as capture the collective interest of traders through search trends, number of web page visitors and/or financial news sentiment. Information emanating from these platforms can significantly affect, or be affected by, changes in the stock market. The overarching goal of this paper is to develop a financial expert system that incorporates these features to predict short term stock prices. Our expert system is comprised of two main modules: a knowledge base and an artificial intelligence (AI) platform. The “knowledge base” for our expert system captures: (a) historical stock prices; (b) several well-known technical indicators; (c) counts and sentiment scores of published news articles for a given stock; (d) trends in Google searches for the given stock ticker; and (e) number of unique visitors for pertinent Wikipedia pages. Once the data is collected, we use a structured approach for data preparation. Then, the AI platform trains four machine learning ensemble methods: (a) a neural network regression ensemble; (b) a support vector regression ensemble; (c) a boosted regression tree; and (d) a random forest regression. In the cross-validation phase, the AI platform picks the “best” ensemble for a given stock. To evaluate the efficacy of our expert system, we first present a case study based on the Citi Group stock ($C) with data collected from 01/01/2013 - 12/31/2016. We show the expert system can predict the 1-day ahead $C stock price with a mean absolute percent error (MAPE) ≤ 1.50% and the 1–10 day ahead with a MAPE ≤ 1.89%, which is better than the reported results in the literature. We show that the use of features extracted from online sources does not substitute the traditional financial metrics, but rather supplements them to improve upon the prediction performance of machine learning based methods. To highlight the utility and generalizability of our expert system, we predict the 1-day ahead price of 19 additional stocks from different industries, volatilities and growth patterns. We report an overall mean for the MAPE statistic of 1.07% across our five different machine learning models, including a MAPE of under 0.75% for 18 of the 19 stocks for the best ensemble (boosted regression tree).
•A financial expert system for predicting the daily stock movements.•“Knowledge base” captures both traditional and online data sources.•The “inference engine” uses three artificial intelligence ...techniques.•Prediction accuracy of 85% is higher than the reported results in the literature.•The system is hosted online and freely available for investors and researchers.
There are several commercial financial expert systems that can be used for trading on the stock exchange. However, their predictions are somewhat limited since they primarily rely on time-series analysis of the market. With the rise of the Internet, new forms of collective intelligence (e.g. Google and Wikipedia) have emerged, representing a new generation of “crowd-sourced” knowledge bases. They collate information on publicly traded companies, while capturing web traffic statistics that reflect the public’s collective interest. Google and Wikipedia have become important “knowledge bases” for investors. In this research, we hypothesize that combining disparate online data sources with traditional time-series and technical indicators for a stock can provide a more effective and intelligent daily trading expert system. Three machine learning models, decision trees, neural networks and support vector machines, serve as the basis for our “inference engine”. To evaluate the performance of our expert system, we present a case study based on the AAPL (Apple NASDAQ) stock. Our expert system had an 85% accuracy in predicting the next-day AAPL stock movement, which outperforms the reported rates in the literature. Our results suggest that: (a) the knowledge base of financial expert systems can benefit from data captured from nontraditional “experts” like Google and Wikipedia; (b) diversifying the knowledge base by combining data from disparate sources can help improve the performance of financial expert systems; and (c) the use of simple machine learning models for inference and rule generation is appropriate with our rich knowledge database. Finally, an intelligent decision making tool is provided to assist investors in making trading decisions on any stock, commodity or index.
Predicting the survival of heart transplant patients is an important, yet challenging problem since it plays a crucial role in understanding the matching procedure between a donor and a recipient. ...Data mining models can be used to effectively analyze and extract novel information from large/complex transplantation datasets. The objective of this study is to predict the 1-, 5-, and 9-year patient's graft survival following a heart transplant surgery via the deployment of analytical models that are based on four powerful classification algorithms (i.e. decision trees, artificial neural networks, support vector machines, and logistic regression). Since the datasets used in this study has a much larger number of survival cases than deaths for 1- and 5-year survival analysis and vice versa for 9-year survival analysis, random under sampling (RUS) and synthetic minority over-sampling (SMOTE) are employed to overcome the data-imbalance problems. The results indicate that logistic regression combined with SMOTE achieves the best classification for the 1-, 5-, and 9-year outcome prediction, with area-under-the-curve (AUC) values of 0.624, 0.676, and 0.838, respectively. By applying sensitivity analysis to the data analytical models, the most important predictors and their associated contribution for the 1-, 5-, and 9-year graft survival of heart transplant patients are identified. By doing so, variables, whose importance changes over time, are differentiated. Not only this proposed hybrid approach gives superior results over the literature but also the models and identification of the variables present important retrospective findings, which can be the basis for a prospective medical study.
•A data-driven approach for predicting survival outcomes at multiple time-points is developed.•The method successfully predicted short-, mid-, & long-term heart transplantation outcomes.•The proposed method is unique in that it fills an important gap in the published literature.•The approach is generic so it can be applied to other organ transplantation cases.
Wearable sensors are currently being used to manage fatigue in professional athletics, transportation and mining industries. In manufacturing, physical fatigue is a challenging ergonomic/safety ...“issue” since it lowers productivity and increases the incidence of accidents. Therefore, physical fatigue must be managed. There are two main goals for this study. First, we examine the use of wearable sensors to detect physical fatigue occurrence in simulated manufacturing tasks. The second goal is to estimate the physical fatigue level over time. In order to achieve these goals, sensory data were recorded for eight healthy participants. Penalized logistic and multiple linear regression models were used for physical fatigue detection and level estimation, respectively. Important features from the five sensors locations were selected using Least Absolute Shrinkage and Selection Operator (LASSO), a popular variable selection methodology. The results show that the LASSO model performed well for both physical fatigue detection and modeling. The modeling approach is not participant and/or workload regime specific and thus can be adopted for other applications.
•A data-driven approach for predicting occurrence and level of the physical fatigue using wearable sensors is developed.•The developed models successfully predicted physical fatigue and non-physical fatigue states.•The proposed set of features are unique in that they are not proposed in the published literature.•The proposed modeling approach is not participant, task or workload regime specific and therefore can be adopted for other applications.
Transcatheter aortic valve replacement (TAVR) is the gold standard treatment for patients with symptomatic aortic stenosis. The utility of existing risk prediction tools for in-hospital mortality ...post-TAVR is limited due to two major factors: (a) the predictive accuracy of these tools is insufficient when only preoperative variables are incorporated, and (b) their efficacy is also compromised when solely postoperative variables are employed, subsequently constraining their application in preoperative decision support. This study examined whether statistical/machine learning models trained with solely preoperative information encoded in the administrative National Inpatient Sample database could accurately predict in-hospital outcomes (death/survival) post-TAVR. Fifteen popular binary classification methods were used to model in-hospital survival/death. These methods were evaluated using multiple classification metrics, including the area under the receiver operating characteristic curve (AUC). By analyzing 54,739 TAVRs, the top five classification models had an AUC ≥ 0.80 for two sampling scenarios: random, consistent with previous studies, and time-based, which assessed whether the models could be deployed without frequent retraining. Given the minimal practical differences in the predictive accuracies of the top five models, the L2 regularized logistic regression model is recommended as the best overall model since it is computationally efficient and easy to interpret.
•Physical fatigue prediction accuracy was ≥ 85% for our two case studies.•Optimizing sensor placement negates the need for multiple sensors.•Heart rate sensor is effective for detecting fatigue in ...supply insertion tasks.•Torso IMU sensor is sufficient for fatigue detection in material handling tasks.•The developed code is freely available for investors and researchers.
The use of expert systems in optimizing and transforming human performance has been limited in practice due to the lack of understanding of how an individual’s performance deteriorates with fatigue accumulation, which can vary based on both the worker and the workplace conditions. As a first step toward realizing the human-centered approach to artificial intelligence and expert systems, this paper lays the foundation for a data analytic approach to managing fatigue in physically-demanding workplaces. The proposed framework capitalizes on continuously collected human performance data from wearable sensor technologies, and is centered around four distinct phases of fatigue: (a) detection, where machine learning methodologies are deployed to detect the occurrence of fatigue; (b) identification, where key features relating to the fatigue occurrence is to be identified; (c) diagnosis, where the fatigue mode is identified based on the knowledge generated in the previous two phases; and (d) recovery, where a suitable intervention is applied to return the worker to mitigate the detrimental effects of fatigue on the worker. Moreover, the framework establishes criteria for feature and machine learning algorithm selection for fatigue management. Two specific application cases of the framework, for two types of manufacturing-related tasks, are presented. Based on the proposed framework and a large number of test sets used in the two case studies, we have shown that: (i) only one wearable sensor is needed for fatigue detection with an average accuracy of ≥ 0.850 and a random forest model comprised of < 7 features; and (ii) the selected features are task-dependent, and thus capturing different modes of fatigue. Therefore, this research presents an important foundation for future expert systems that attempt to quantify/predict changes in workers’ performance as an input to prescriptive rest-break scheduling, job-rotation, and task assignment models. To encourage future work in this important area, we provide links to our data and code as Supplementary materials.
The purpose of this study is to provide a method for classifying non-fatigued vs. fatigued states following manual material handling. A method of template matching pattern recognition for feature ...extraction ($1 Recognizer) along with the support vector machine model for classification were applied on the kinematics of gait cycles segmented by our stepwise search-based segmentation algorithm. A single inertial measurement unit on the ankle was used, providing a minimally intrusive and inexpensive tool for monitoring. The classifier distinguished between states using distance-based scores from the recogniser and the step duration. The results of fatigue detection showed an accuracy of 90% across data from 20 recruited subjects. This method utilises the minimum amount of data and features from only one low-cost sensor to reliably classify the state of fatigue induced by a realistic manufacturing task using a simple machine learning algorithm that can be extended to real-time fatigue monitoring as a future technology to be employed in the manufacturing facilities.
Practitioner Summary: We examined the use of a wearable sensor for the detection of fatigue-related changes in gait based on a simulated manual material handling task. Classification based on foot acceleration and position trajectories resulted in 90% accuracy. This method provides a practical framework for predicting realistic levels of fatigue.
Recent research has shown that data mining models can accurately predict the outcome of a heart transplant based on predictors that include patient and donor's health/demographics. These models have ...not been adopted in practice, however, since they did not: a) consider the interactions between the explanatory variables; b) provide a patient's specific risk of survival (reported results have been primarily deterministic); and c) offer an automated decision tool that can provide some data-driven insights to practitioners. In this study, we attempt to overcome these three limitations through the use of Bayesian Belief Networks (BBN). The proposed BBN framework is comprised of four phases. In the first two phases, the data is preprocessed, and a candidate set of predictors is generated based on employing several variable selection methods. The third phase involves the addition of medically relevant variables to the list. In phase four, the BBN model is applied. The results show that the proposed BBN method provides similar predictive performance to the best approaches in the literature. More importantly, our method provides novel information on the interactions among the predictors and the conditional probability of survival for a given set of relevant donor–recipient characteristics. We offer U.S. practitioners a decision support tool that presents an individualized survival score based on our BBN model (and the UNOS dataset).
•A model that provides competitive predictors for 9-year heart transplantation outcomes.•The BBN model presents data-driven insights into how the predictors interact.•The result of the methodology is a novel individualized risk score for heart transplant outcomes.•A decision support tool is provided to assist practitioners in characterizing a heart transplant's risk.
Objective The COVID-19 pandemic in the U.S. has exhibited a distinct multiwave pattern beginning in March 2020. Paradoxically, most counties do not exhibit this same multiwave pattern. We aim to ...answer three research questions: (1) How many distinct clusters of counties exhibit similar COVID-19 patterns in the time-series of daily confirmed cases? (2) What is the geographic distribution of the counties within each cluster? and (3) Are county-level demographic, socioeconomic and political variables associated with the COVID-19 case patterns? Materials and methods We analyzed data from counties in the U.S. from March 1, 2020 to January 2, 2021. Time series clustering identified clusters in the daily confirmed cases of COVID-19. An explanatory model was used to identify demographic, socioeconomic and political variables associated with the outbreak patterns. Results Three patterns were identified from the cluster solution including counties in which cases are still increasing, those that peaked in the late fall, and those with low case counts to date. Several county-level demographic, socioeconomic, and political variables showed significant associations with the identified clusters. Discussion The pattern of the outbreak is related both to the geographic location within the U.S. and several variables including population density and government response. Conclusion The reported pattern of cases in the U.S. is observed through aggregation of the daily confirmed COVID-19 cases, suggesting that local trends may be more informative. The pattern of the outbreak varies by county, and is associated with important demographic, socioeconomic, political and geographic factors.
PDF
