Search results for: LSTM prediction model

Authors: Juan Sebastián Hernández

Abstract:

The Nevado del Ruíz volcano, located between the limits of the Departments of Caldas and Tolima in Colombia, presented an unstable behaviour in the course of the year 2020, this volcanic activity led to secondary effects on the crust, which is why the prediction of deformations becomes the task of geoscientists. In the course of this article, the use of tropospheric variables such as evapotranspiration, UV aerosol index, carbon monoxide, nitrogen dioxide, methane, surface temperature, among others, is used to train a set of neural networks that can predict the behaviour of the resulting phase of an unrolled interferogram with the DInSAR technique, whose main objective is to identify and characterise the behaviour of the crust based on the environmental conditions. For this purpose, variables were collected, a generalised linear model was created, and a set of neural networks was created. After the training of the network, validation was carried out with the test data, giving an MSE of 0.17598 and an associated r-squared of approximately 0.88454. The resulting model provided a dataset with good thematic accuracy, reflecting the behaviour of the volcano in 2020, given a set of environmental characteristics.

Keywords: crustal deformation, Tropomi, neural networks (ANN), volcanic activity, DInSAR

Procedia PDF Downloads 76

Authors: Natalia Feruleva

Abstract:

The goal of this paper is to develop the fraud risk assessment model basing on both relational and financial data and test the impact of the relationships between Russian non-public companies on the likelihood of financial fraud commitment. Relationships mean various linkages between companies such as parent-subsidiary relationship and person-related relationships. These linkages may provide additional opportunities for committing fraud. Person-related relationships appear when firms share a director, or the director owns another firm. The number of companies belongs to CEO and managed by CEO, the number of subsidiaries was calculated to measure the relationships. Moreover, the dummy variable describing the existence of parent company was also included in model. Control variables such as financial leverage and return on assets were also implemented because they describe the motivating factors of fraud. To check the hypotheses about the influence of the chosen parameters on the likelihood of financial fraud, information about person-related relationships between companies, existence of parent company and subsidiaries, profitability and the level of debt was collected. The resulting sample consists of 160 Russian non-public firms. The sample includes 80 fraudsters and 80 non-fraudsters operating in 2006-2017. The dependent variable is dichotomous, and it takes the value 1 if the firm is engaged in financial crime, otherwise 0. Employing probit model, it was revealed that the number of companies which belong to CEO of the firm or managed by CEO has significant impact on the likelihood of financial fraud. The results obtained indicate that the more companies are affiliated with the CEO, the higher the likelihood that the company will be involved in financial crime. The forecast accuracy of the model is about is 80%. Thus, the model basing on both relational and financial data gives high level of forecast accuracy.

Keywords: financial fraud, fraud prediction, non-public companies, regression analysis, relational data

Procedia PDF Downloads 101

Authors: Trupti Diwan

Abstract:

Predicting student failure at school has changed into a difficult challenge due to both the large number of factors that can affect the reduced performance of students and the imbalanced nature of these kinds of data sets. This paper surveys the two elements needed to make prediction on Students’ Academic Performances which are parameters and methods. This paper also proposes a framework for predicting the performance of engineering students. Genetic programming can be used to predict student failure/success. Ranking algorithm is used to rank students according to their credit points. The framework can be used as a basis for the system implementation & prediction of students’ Academic Performance in Higher Learning Institute.

Keywords: classification, educational data mining, student failure, grammar-based genetic programming

Procedia PDF Downloads 403

Authors: Haiyan Wu, Ying Liu, Shaoyun Shi

Abstract:

Authorship attribution is to extract features to identify authors of anonymous documents. Many previous works on authorship attribution focus on statistical style features (e.g., sentence/word length), content features (e.g., frequent words, n-grams). Modeling these features by regression or some transparent machine learning methods gives a portrait of the authors' writing style. But these methods do not capture the syntactic (e.g., dependency relationship) or semantic (e.g., topics) information. In recent years, some researchers model syntactic trees or latent semantic information by neural networks. However, few works take them together. Besides, predictions by neural networks are difficult to explain, which is vital in authorship attribution tasks. In this paper, we not only utilize the statistical style and content features but also take advantage of both syntactic and semantic features. Different from an end-to-end neural model, feature selection and prediction are two steps in our method. An attentive n-gram network is utilized to select useful features, and logistic regression is applied to give prediction and understandable representation of writing style. Experiments show that our extracted features can improve the state-of-the-art methods on three benchmark datasets.

Keywords: authorship attribution, attention mechanism, syntactic feature, feature extraction

Procedia PDF Downloads 115

Authors: Misuk Kim, Seokho Kang, Jehyuk Lee, Hyunchang Cho, Sungzoon Cho

Abstract:

In semiconductor manufacturing, virtual metrology (VM) refers to methods to predict properties of a wafer based on machine parameters and sensor data of the production equipment, without performing the (costly) physical measurement of the wafer properties (Wikipedia). Additional benefits include avoidance of human bias and identification of important factors affecting the quality of the process which allow improving the process quality in the future. It is however rare to find VM applied to other areas of manufacturing. In this work, we propose to use VM to copper clad laminate (CCL) manufacturing. CCL is a core element of a printed circuit board (PCB) which is used in smartphones, tablets, digital cameras, and laptop computers. The manufacturing of CCL consists of three processes: Treating, lay-up, and pressing. Treating, the most important process among the three, puts resin on glass cloth, heat up in a drying oven, then produces prepreg for lay-up process. In this process, three important quality factors are inspected: Treated weight (T/W), Minimum Viscosity (M/V), and Gel Time (G/T). They are manually inspected, incurring heavy cost in terms of time and money, which makes it a good candidate for VM application. We developed prediction models of the three quality factors T/W, M/V, and G/T, respectively, with process variables, raw material, and environment variables. The actual process data was obtained from a CCL manufacturer. A variety of variable selection methods and learning algorithms were employed to find the best prediction model. We obtained prediction models of M/V and G/T with a high enough accuracy. They also provided us with information on “important” predictor variables, some of which the process engineers had been already aware and the rest of which they had not. They were quite excited to find new insights that the model revealed and set out to do further analysis on them to gain process control implications. T/W did not turn out to be possible to predict with a reasonable accuracy with given factors. The very fact indicates that the factors currently monitored may not affect T/W, thus an effort has to be made to find other factors which are not currently monitored in order to understand the process better and improve the quality of it. In conclusion, VM application to CCL’s treating process was quite successful. The newly built quality prediction model allowed one to reduce the cost associated with actual metrology as well as reveal some insights on the factors affecting the important quality factors and on the level of our less than perfect understanding of the treating process.

Keywords: copper clad laminate, predictive modeling, quality control, virtual metrology

Procedia PDF Downloads 336

Authors: Smail Tigani, Mohamed Ouzzif

Abstract:

This work presents a discrete quantitative state prediction algorithm with intelligent behavior making it able to self-improve some performance aspects. The specificity of this algorithm is the capacity of self-rectification of the prediction strategy before the final decision. The auto-rectification mechanism is based on two parallel mathematical models. In one hand, the algorithm predicts the next state based on event transition matrix updated after each observation. In the other hand, the algorithm extracts its residues trend with a linear regression representing historical residues data-points in order to rectify the first decision if needs. For a normal distribution, the interactivity between the two models allows the algorithm to self-optimize its performance and then make better prediction. Designed key performance indicator, computed during a Monte Carlo simulation, shows the advantages of the proposed approach compared with traditional one.

Keywords: discrete state, Markov Chains, linear regression, auto-adaptive systems, decision making, Monte Carlo Simulation

Procedia PDF Downloads 480

Authors: U. Chattaraj, K. Dhusiya, M. Raviteja

Abstract:

Perception reaction time of drivers is an outcome of human thought process, which is vague and approximate in nature and also varies from driver to driver. So, in this study a fuzzy logic based model for prediction of the same has been presented, which seems suitable. The control factors, like, age, experience, intensity of driving of the driver, speed of the vehicle and distance of stimulus have been considered as premise variables in the model, in which the perception reaction time is the consequence variable. Results show that the model is able to explain the impacts of the control factors on perception reaction time properly.

Keywords: driver, fuzzy logic, perception reaction time, premise variable

Procedia PDF Downloads 277

Authors: Piotr A. Kowalski, Kasper Sapala, Wiktor Warchalowski

Abstract:

PM10 is a suspended dust that primarily has a negative effect on the respiratory system. PM10 is responsible for attacks of coughing and wheezing, asthma or acute, violent bronchitis. Indirectly, PM10 also negatively affects the rest of the body, including increasing the risk of heart attack and stroke. Unfortunately, Poland is a country that cannot boast of good air quality, in particular, due to large PM concentration levels. Therefore, based on the dense network of Airly sensors, it was decided to deal with the problem of prediction of suspended particulate matter concentration. Due to the very complicated nature of this issue, the Machine Learning approach was used. For this purpose, Convolution Neural Network (CNN) neural networks have been adopted, these currently being the leading information processing methods in the field of computational intelligence. The aim of this research is to show the influence of particular CNN network parameters on the quality of the obtained forecast. The forecast itself is made on the basis of parameters measured by Airly sensors and is carried out for the subsequent day, hour after hour. The evaluation of learning process for the investigated models was mostly based upon the mean square error criterion; however, during the model validation, a number of other methods of quantitative evaluation were taken into account. The presented model of pollution prediction has been verified by way of real weather and air pollution data taken from the Airly sensor network. The dense and distributed network of Airly measurement devices enables access to current and archival data on air pollution, temperature, suspended particulate matter PM1.0, PM2.5, and PM10, CAQI levels, as well as atmospheric pressure and air humidity. In this investigation, PM2.5, and PM10, temperature and wind information, as well as external forecasts of temperature and wind for next 24h served as inputted data. Due to the specificity of the CNN type network, this data is transformed into tensors and then processed. This network consists of an input layer, an output layer, and many hidden layers. In the hidden layers, convolutional and pooling operations are performed. The output of this system is a vector containing 24 elements that contain prediction of PM10 concentration for the upcoming 24 hour period. Over 1000 models based on CNN methodology were tested during the study. During the research, several were selected out that give the best results, and then a comparison was made with the other models based on linear regression. The numerical tests carried out fully confirmed the positive properties of the presented method. These were carried out using real ‘big’ data. Models based on the CNN technique allow prediction of PM10 dust concentration with a much smaller mean square error than currently used methods based on linear regression. What's more, the use of neural networks increased Pearson's correlation coefficient (R²) by about 5 percent compared to the linear model. During the simulation, the R² coefficient was 0.92, 0.76, 0.75, 0.73, and 0.73 for 1st, 6th, 12th, 18th, and 24th hour of prediction respectively.

Keywords: air pollution prediction (forecasting), machine learning, regression task, convolution neural networks

Procedia PDF Downloads 122

Authors: Elison Eduardo Jardim Bierhals, Claudineia Brazil, Deivid Pires, Rafael Haag, Elton Gimenez Rossini

Abstract:

This study evaluated the uncertainties involved in the solar radiation projections generated by the Brazilian Earth System Model (BESM) of the Weather and Climate Prediction Center (CPTEC) belonging to Coupled Model Intercomparison Phase 5 (CMIP5), with the aim of identifying efficiency in the projections for solar radiation of said model and in this way establish the viability of its use. Two different scenarios elaborated by Intergovernmental Panel on Climate Change (IPCC) were evaluated: RCP 4.5 (with more optimistic contour conditions) and 8.5 (with more pessimistic initial conditions). The method used to verify the accuracy of the present model was the Nash coefficient and the Statistical bias, as it better represents these atmospheric patterns. The BESM showed a tendency to overestimate the data ​​of solar radiation projections in most regions of the state of Rio Grande do Sul and through the validation methods adopted by this study, BESM did not present a satisfactory accuracy.

Keywords: climate changes, projections, solar radiation, uncertainty

Procedia PDF Downloads 226

Authors: Belkacem Selma, Boumediene Selma, Tourkia Guerzou, Abbes Labdelli

Abstract:

Natural disasters have occurred and will continue to cause human and material damage. Therefore, the idea of "preventing" natural disasters will never be possible. However, their prediction is possible with the advancement of technology. Even if natural disasters are effectively inevitable, their consequences may be partly controlled. The rapid growth and progress of artificial intelligence (AI) had a major impact on the prediction of natural disasters and risk assessment which are necessary for effective disaster reduction. The Earthquakes prediction to prevent the loss of human lives and even property damage is an important factor; that is why it is crucial to develop techniques for predicting this natural disaster. This present study aims to analyze the ability of artificial neural networks (ANNs) to predict earthquakes that occur in a given area. The used data describe the problem of high energy (higher than 10^4J) seismic bumps forecasting in a coal mine using two long walls as an example. For this purpose, seismic bumps data obtained from mines has been analyzed. The results obtained show that the ANN with high accuracy was able to predict earthquake parameters; the classification accuracy through neural networks is more than 94%, and that the models developed are efficient and robust and depend only weakly on the initial database.

Keywords: earthquake prediction, ANN, seismic bumps

Procedia PDF Downloads 106

Authors: The Danh Phan

Abstract:

House price forecasting is a main topic in the real estate market research. Effective house price prediction models could not only allow home buyers and real estate agents to make better data-driven decisions but may also be beneficial for the property policymaking process. This study investigates the housing market by using machine learning techniques to analyze real historical house sale transactions in Australia. It seeks useful models which could be deployed as an application for house buyers and sellers. Data analytics show a high discrepancy between the house price in the most expensive suburbs and the most affordable suburbs in the city of Melbourne. In addition, experiments demonstrate that the combination of Stepwise and Support Vector Machine (SVM), based on the Mean Squared Error (MSE) measurement, consistently outperforms other models in terms of prediction accuracy.

Keywords: house price prediction, regression trees, neural network, support vector machine, stepwise

Procedia PDF Downloads 199

Authors: Luis Fonseca, Filipe Cabral Pinto, Susana Sargento

Abstract:

The increase of the world population, especially in large urban centers, has resulted in new challenges particularly with the control and optimization of public safety. Thus, in the present work, a solution is proposed for the prediction of criminal occurrences in a city based on historical data of incidents and demographic information. The entire research and implementation will be presented start with the data collection from its original source, the treatment and transformations applied to them, choice and the evaluation and implementation of the Machine Learning model up to the application layer. Classification models will be implemented to predict criminal risk for a given time interval and location. Machine Learning algorithms such as Random Forest, Neural Networks, K-Nearest Neighbors and Logistic Regression will be used to predict occurrences, and their performance will be compared according to the data processing and transformation used. The results show that the use of Machine Learning techniques helps to anticipate criminal occurrences, which contributed to the reinforcement of public security. Finally, the models were implemented on a platform that will provide an API to enable other entities to make requests for predictions in real-time. An application will also be presented where it is possible to show criminal predictions visually.

Keywords: crime prediction, machine learning, public safety, smart city

Procedia PDF Downloads 89

Authors: C. S. Subhashini, H. L. Premaratne

Abstract:

Landslides are the most recurrent and prominent disaster in Sri Lanka. Sri Lanka has been subjected to a number of extreme landslide disasters that resulted in a significant loss of life, material damage, and distress. It is required to explore a solution towards preparedness and mitigation to reduce recurrent losses associated with landslides. Artificial Neural Networks (ANNs) and Hidden Markov Model (HMMs) are now widely used in many computer applications spanning multiple domains. This research examines the effectiveness of using Artificial Neural Networks and Hidden Markov Model in landslides predictions and the possibility of applying the modern technology to predict landslides in a prominent geographical area in Sri Lanka. A thorough survey was conducted with the participation of resource persons from several national universities in Sri Lanka to identify and rank the influencing factors for landslides. A landslide database was created using existing topographic; soil, drainage, land cover maps and historical data. The landslide related factors which include external factors (Rainfall and Number of Previous Occurrences) and internal factors (Soil Material, Geology, Land Use, Curvature, Soil Texture, Slope, Aspect, Soil Drainage, and Soil Effective Thickness) are extracted from the landslide database. These factors are used to recognize the possibility to occur landslides by using an ANN and HMM. The model acquires the relationship between the factors of landslide and its hazard index during the training session. These models with landslide related factors as the inputs will be trained to predict three classes namely, ‘landslide occurs’, ‘landslide does not occur’ and ‘landslide likely to occur’. Once trained, the models will be able to predict the most likely class for the prevailing data. Finally compared two models with regards to prediction accuracy, False Acceptance Rates and False Rejection rates and This research indicates that the Artificial Neural Network could be used as a strong decision support system to predict landslides efficiently and effectively than Hidden Markov Model.

Keywords: landslides, influencing factors, neural network model, hidden markov model

Procedia PDF Downloads 364

Authors: Ji-Seok Koo, Hee‑Yong Kwon, Hui-Young Yun, Kyung-Hui Wang, Youn-Seo Koo

Abstract:

This paper presents a forecasting system for PM2.5 levels in Seoul, South Korea, leveraging a combination of chemical transport modeling and ConvLSTM-DNN machine learning technology. Exposure to PM2.5 has known detrimental impacts on public health, making its prediction crucial for establishing preventive measures. Existing forecasting models, like the Community Multiscale Air Quality (CMAQ) and Weather Research and Forecasting (WRF), are hindered by their reliance on uncertain input data, such as anthropogenic emissions and meteorological patterns, as well as certain intrinsic model limitations. The system we've developed specifically addresses these issues by integrating machine learning and using carefully selected input features that account for local and distant sources of PM2.5. In South Korea, the PM2.5 concentration is greatly influenced by both local emissions and long-range transport from China, and our model effectively captures these spatial and temporal dynamics. Our PM2.5 prediction system combines the strengths of advanced hybrid machine learning algorithms, convLSTM and DNN, to improve upon the limitations of the traditional CMAQ model. Data used in the system include forecasted information from CMAQ and WRF models, along with actual PM2.5 concentration and weather variable data from monitoring stations in China and South Korea. The system was implemented specifically for Seoul's PM2.5 forecasting.

Keywords: PM2.5 forecast, machine learning, convLSTM, DNN

Procedia PDF Downloads 40

Authors: Yong Zhao, Jian He, Cheng Zhang

Abstract:

Cardiovascular diseases caused by hypertension are extremely threatening to human health, and early diagnosis of hypertension can save a large number of lives. Traditional hypertension detection methods require special equipment and are difficult to detect continuous blood pressure changes. In this regard, this paper first analyzes the principle of heart rate variability (HRV) and introduces sliding window and power spectral density (PSD) to analyze the time domain features and frequency domain features of HRV, and secondly, designs an HRV-based hypertension prediction network by combining Resnet, attention mechanism, and multilayer perceptron, which extracts the frequency domain through the improved ResNet18 features through a modified ResNet18, its fusion with time-domain features through an attention mechanism, and the auxiliary prediction of hypertension through a multilayer perceptron. Finally, the network was trained and tested using the publicly available SHAREE dataset on PhysioNet, and the test results showed that this network achieved 92.06% prediction accuracy for hypertension and outperformed K Near Neighbor(KNN), Bayes, Logistic, and traditional Convolutional Neural Network(CNN) models in prediction performance.

Keywords: feature extraction, heart rate variability, hypertension, residual networks

Procedia PDF Downloads 73

Authors: Ben Clapperton, Daniel Stahl, Kimberley Goldsmith, Trudie Chalder

Abstract:

Chronic fatigue syndrome (CFS) is a condition characterised by chronic disabling fatigue and other symptoms that currently can't be explained by any underlying medical condition. Although clinical trials support the effectiveness of cognitive behaviour therapy (CBT), the success rate for individual patients is modest. Patients vary in their response and little is known which factors predict or moderate treatment outcomes. The aim of the project is to develop a prediction model from baseline characteristics of patients, such as demographics, clinical and psychological variables, which may predict likely treatment outcome and provide guidance for clinical decision making and help clinicians to recommend the best treatment. The project is aimed at identifying subgroups of patients with similar baseline characteristics that are predictive of treatment effects using modern cluster analyses and data mining machine learning algorithms. The characteristics of these groups will then be used to inform the types of individuals who benefit from a specific treatment. In addition, results will provide a better understanding of for whom the treatment works. The suitability of different clustering methods to identify subgroups and their response to different treatments of CFS patients is compared.

Keywords: chronic fatigue syndrome, latent class analysis, prediction modelling, self-organizing maps

Procedia PDF Downloads 202

Authors: Amadou Wurry Jallow, Adama N. S. Bah, Karamo Bah, Shih-Ye Wang, Kuo-Chung Chu, Chien-Yeh Hsu

Abstract:

Chronic kidney disease (CKD) is a major public health challenge with high prevalence, rising incidence, and serious adverse consequences. Developing effective risk prediction models is a cost-effective approach to predicting and preventing complications of chronic kidney disease (CKD). This study aimed to develop an accurate machine learning model that can dynamically identify individuals at risk of CKD using various kinds of diagnostic data, with or without laboratory data, at different follow-up points. Creatinine is a key component used to predict CKD. These models will enable affordable and effective screening for CKD even with incomplete patient data, such as the absence of creatinine testing. This retrospective cohort study included data on 19,429 adults provided by a private research institute and screening laboratory in Taiwan, gathered between 2001 and 2015. Univariate Cox proportional hazard regression analyses were performed to determine the variables with high prognostic values for predicting CKD. We then identified interacting variables and grouped them according to diagnostic data categories. Our models used three types of data gathered at three points in time: non-laboratory, laboratory, and metabolic indices data. Next, we used subgroups of variables within each category to train two machine learning models (Random Forest and XGBoost). Our machine learning models can dynamically discriminate individuals at risk for developing CKD. All the models performed well using all three kinds of data, with or without laboratory data. Using only non-laboratory-based data (such as age, sex, body mass index (BMI), and waist circumference), both models predict chronic kidney disease as accurately as models using laboratory and metabolic indices data. Our machine learning models have demonstrated the use of different categories of diagnostic data for CKD prediction, with or without laboratory data. The machine learning models are simple to use and flexible because they work even with incomplete data and can be applied in any clinical setting, including settings where laboratory data is difficult to obtain.

Keywords: chronic kidney disease, glomerular filtration rate, creatinine, novel metabolic indices, machine learning, risk prediction

Procedia PDF Downloads 81

Authors: Leonardo Juan Ramírez López, Javier Oswaldo Rodriguez Velasquez

Abstract:

We have designed a Holter that measures the heart´s activity for over 24 hours, implemented a prediction methodology, and generate alarms as well as indicators to patients and treating physicians. Various diagnostic advances have been developed in clinical cardiology thanks to Holter implementation; however, their interpretation has largely been conditioned to clinical analysis and measurements adjusted to diverse population characteristics, thus turning it into a subjective examination. This, however, requires vast population studies to be validated that, in turn, have not achieved the ultimate goal: mortality prediction. Given this context, our Insight Research Group developed a mathematical methodology that assesses cardiac dynamics through entropy and probability, creating a numerical and geometrical attractor which allows quantifying the normalcy of chronic and acute disease as well as the evolution between such states, and our Tigum Research Group developed a holter device with 12 channels and advanced computer software. This has been shown in different contexts with 100% sensitivity and specificity results.

Keywords: attractor , cardiac, entropy, holter, mathematical , prediction

Procedia PDF Downloads 149

Authors: Shenlun Chen, Leonard Wee

Abstract:

Tumor grading is an essential reference for colorectal cancer (CRC) staging and survival prognostication. The widely used World Health Organization (WHO) grading system defines histological grade of CRC adenocarcinoma based on the density of glandular formation on whole slide images (WSI). Tumors are classified as well-, moderately-, poorly- or un-differentiated depending on the percentage of the tumor that is gland forming; >95%, 50-95%, 5-50% and <5%, respectively. However, manually grading WSIs is a time-consuming process and can cause observer error due to subjective judgment and unnoticed regions. Furthermore, pathologists’ grading is usually coarse while a finer and continuous differentiation grade may help to stratifying CRC patients better. In this study, a deep learning based automatic differentiation grading algorithm was developed and evaluated by survival analysis. Firstly, a gland segmentation model was developed for segmenting gland structures. Gland regions of WSIs were delineated and used for differentiation annotating. Tumor regions were annotated by experienced pathologists into high-, medium-, low-differentiation and normal tissue, which correspond to tumor with clear-, unclear-, no-gland structure and non-tumor, respectively. Then a differentiation prediction model was developed on these human annotations. Finally, all enrolled WSIs were processed by gland segmentation model and differentiation prediction model. The differentiation grade can be calculated by deep learning models’ prediction of tumor regions and tumor differentiation status according to WHO’s defines. If multiple WSIs were possessed by a patient, the highest differentiation grade was chosen. Additionally, the differentiation grade was normalized into scale between 0 to 1. The Cancer Genome Atlas, project COAD (TCGA-COAD) project was enrolled into this study. For the gland segmentation model, receiver operating characteristic (ROC) reached 0.981 and accuracy reached 0.932 in validation set. For the differentiation prediction model, ROC reached 0.983, 0.963, 0.963, 0.981 and accuracy reached 0.880, 0.923, 0.668, 0.881 for groups of low-, medium-, high-differentiation and normal tissue in validation set. Four hundred and one patients were selected after removing WSIs without gland regions and patients without follow up data. The concordance index reached to 0.609. Optimized cut off point of 51% was found by “Maxstat” method which was almost the same as WHO system’s cut off point of 50%. Both WHO system’s cut off point and optimized cut off point performed impressively in Kaplan-Meier curves and both p value of logrank test were below 0.005. In this study, gland structure of WSIs and differentiation status of tumor regions were proven to be predictable through deep leaning method. A finer and continuous differentiation grade can also be automatically calculated through above models. The differentiation grade was proven to stratify CAC patients well in survival analysis, whose optimized cut off point was almost the same as WHO tumor grading system. The tool of automatically calculating differentiation grade may show potential in field of therapy decision making and personalized treatment.

Keywords: colorectal cancer, differentiation, survival analysis, tumor grading

Procedia PDF Downloads 120

Authors: Afroze Ibrahim Baqapuri, Ilya Trofimov

Abstract:

Sponsored search is a multi-billion dollar industry and makes up a major source of revenue for search engines (SE). Click-through-rate (CTR) estimation plays a crucial role for ads selection, and greatly affects the SE revenue, advertiser traffic and user experience. We propose a novel architecture of solving CTR prediction problem by combining artificial neural networks (ANN) with decision trees. First, we compare ANN with respect to other popular machine learning models being used for this task. Then we go on to combine ANN with MatrixNet (proprietary implementation of boosted trees) and evaluate the performance of the system as a whole. The results show that our approach provides a significant improvement over existing models.

Keywords: neural networks, sponsored search, web advertisement, click prediction, click-through rate

Procedia PDF Downloads 552

Authors: Welawat Tienpratarn, Chaiyaporn Yuksen, Rungrawin Promkul, Chetsadakon Jenpanitpong, Pajit Bunta, Suthap Jaiboon

Abstract:

Supraventricular tachycardia (SVT) is an abnormally fast atrial tachycardia characterized by narrow (≤ 120 ms) and constant QRS. Adenosine was the drug of choice; the first dose was 6 mg. It can be repeated with the second and third doses of 12 mg, with greater than 90% success. The study found that patients observed at 4 hours after normal sinus rhythm was no recurrence within 24 hours. The objective of this study was to investigate the factors that influence the recurrence of SVT after adenosine in the emergency department (ED). The study was conducted retrospectively exploratory model, prognostic study at the Emergency Department (ED) in Faculty of Medicine, Ramathibodi Hospital, a university-affiliated super tertiary care hospital in Bangkok, Thailand. The study was conducted for ten years period between 2010 and 2020. The inclusion criteria were age > 15 years, visiting the ED with SVT, and treating with adenosine. Those patients were recorded with the recurrence SVT in ED. The multivariable logistic regression model developed the predictive model and prediction score for recurrence PSVT. 264 patients met the study criteria. Of those, 24 patients (10%) had recurrence PSVT. Five independent factors were predictive of recurrence PSVT. There was age>65 years, heart rate (after adenosine) > 100 per min, structural heart disease, and dose of adenosine. The clinical risk score to predict recurrence PSVT is developed accuracy 74.41%. The score of >6 had the likelihood ratio of recurrence PSVT by 5.71 times. The clinical predictive score of > 6 was associated with recurrence PSVT in ED.

Keywords: supraventricular tachycardia, recurrance, emergency department, adenosine

Procedia PDF Downloads 93

Authors: Sakshi Dhumale, Madhushree C., Amba Shetty

Abstract:

The effect of station density on the geostatistical prediction of groundwater levels is of critical importance to ensure accurate and reliable predictions. Monitoring station density directly impacts the accuracy and reliability of geostatistical predictions by influencing the model's ability to capture localized variations and small-scale features in groundwater levels. This is particularly crucial in regions with complex hydrogeological conditions and significant spatial heterogeneity. Insufficient station density can result in larger prediction uncertainties, as the model may struggle to adequately represent the spatial variability and correlation patterns of the data. On the other hand, an optimal distribution of monitoring stations enables effective coverage of the study area and captures the spatial variability of groundwater levels more comprehensively. In this study, we investigate the effect of station density on the predictive performance of groundwater levels using the geostatistical technique of Ordinary Kriging. The research utilizes groundwater level data collected from 121 observation wells within the semi-arid Berambadi watershed, gathered over a six-year period (2010-2015) from the Indian Institute of Science (IISc), Bengaluru. The dataset is partitioned into seven subsets representing varying sampling densities, ranging from 15% (12 wells) to 100% (121 wells) of the total well network. The results obtained from different monitoring networks are compared against the existing groundwater monitoring network established by the Central Ground Water Board (CGWB). The findings of this study demonstrate that higher station densities significantly enhance the accuracy of geostatistical predictions for groundwater levels. The increased number of monitoring stations enables improved interpolation accuracy and captures finer-scale variations in groundwater levels. These results shed light on the relationship between station density and the geostatistical prediction of groundwater levels, emphasizing the importance of appropriate station densities to ensure accurate and reliable predictions. The insights gained from this study have practical implications for designing and optimizing monitoring networks, facilitating effective groundwater level assessments, and enabling sustainable management of groundwater resources.

Keywords: station density, geostatistical prediction, groundwater levels, monitoring networks, interpolation accuracy, spatial variability

Procedia PDF Downloads 32

Authors: Vasujeet Singh, Pruthiviraj Nemalipuri, Vivek Vitankar, Harish Chandra Das

Abstract:

For the correct prediction of thermal and hydraulic performance (bed voidage, suspension density, pressure drop, heat transfer, and combustion kinetics), one should incorporate the correct parameters in the computational fluid dynamics simulation of a fluidized bed gasifier. Scarcity of fossil fuels, and to fulfill the energy demand of the increasing population, researchers need to shift their attention to the alternative to fossil fuels. The current research work focuses on hydrodynamics behavior and gasification of sawdust inside a 2D industrial scale FBG using the Eulerian-Eulerian multifluid model. The present numerical model is validated with experimental data. Further, this model extended for the prediction of gasification characteristics of sawdust by incorporating eight heterogeneous moisture release, volatile cracking, tar cracking, tar oxidation, char combustion, CO₂ gasification, steam gasification, methanation reaction, and five homogeneous oxidation of CO, CH₄, H₂, forward and backward water gas shift (WGS) reactions. In the result section, composition of gasification products is analyzed, along with the hydrodynamics of sawdust and sand phase, heat transfer between the gas, sand and sawdust, reaction rates of different homogeneous and heterogeneous reactions is being analyzed along the height of the domain.

Keywords: devolatilization, Eulerian-Eulerian, fluidized bed gasifier, mathematical modelling, sawdust gasification

Procedia PDF Downloads 87

Authors: Rachel Y. Zhang, Christopher K. Anderson

Abstract:

A critical aspect of revenue management is a firm’s ability to predict demand as a function of price. Historically hotels have used simple time series models (regression and/or pick-up based models) owing to the complexities of trying to build casual models of demands. Machine learning approaches are slowly attracting attention owing to their flexibility in modeling relationships. This study provides an overview of approaches to forecasting hospitality demand – focusing on the opportunities created by machine learning approaches, including K-Nearest-Neighbors, Support vector machine, Regression Tree, and Artificial Neural Network algorithms. The out-of-sample performances of above approaches to forecasting hotel demand are illustrated by using a proprietary sample of the market level (24 properties) transactional data for Las Vegas NV. Causal predictive models can be built and evaluated owing to the availability of market level (versus firm level) data. This research also compares and contrast model accuracy of firm-level models (i.e. predictive models for hotel A only using hotel A’s data) to models using market level data (prices, review scores, location, chain scale, etc… for all hotels within the market). The prospected models will be valuable for hotel revenue prediction given the basic characters of a hotel property or can be applied in performance evaluation for an existed hotel. The findings will unveil the features that play key roles in a hotel’s revenue performance, which would have considerable potential usefulness in both revenue prediction and evaluation.

Keywords: hotel revenue, k-nearest-neighbors, machine learning, neural network, prediction model, regression tree, support vector machine

Procedia PDF Downloads 114

Authors: Dongping Fang, Lian Duan, Xiaojing Yuan, Mike Xu, Allyn Klunder, Kevin Tan, Suiting Cao, Yeqing Ji

Abstract:

Accurate prediction of a medical condition with straight clinical evidence is a long-sought topic in the medical management and health insurance field. Although great progress has been made with machine learning algorithms, the medical community is still, to a certain degree, suspicious about the model's accuracy and interpretability. This paper presents an innovative hierarchical attention deep learning model to achieve good prediction and clear interpretability that can be easily understood by medical professionals. This deep learning model uses a hierarchical attention structure that matches naturally with the medical history data structure and reflects the member’s encounter (date of service) sequence. The model attention structure consists of 3 levels: (1) attention on the medical code types (diagnosis codes, procedure codes, lab test results, and prescription drugs), (2) attention on the sequential medical encounters within a type, (3) attention on the medical codes within an encounter and type. This model is applied to predict the occurrence of stage 3 chronic kidney disease (CKD3), using three years’ medical history of Medicare Advantage (MA) members from a top health insurance company. The model takes members’ medical events, both claims and electronic medical record (EMR) data, as input, makes a prediction of CKD3 and calculates the contribution from individual events to the predicted outcome. The model outcome can be easily explained with the clinical evidence identified by the model algorithm. Here are examples: Member A had 36 medical encounters in the past three years: multiple office visits, lab tests and medications. The model predicts member A has a high risk of CKD3 with the following well-contributed clinical events - multiple high ‘Creatinine in Serum or Plasma’ tests and multiple low kidneys functioning ‘Glomerular filtration rate’ tests. Among the abnormal lab tests, more recent results contributed more to the prediction. The model also indicates regular office visits, no abnormal findings of medical examinations, and taking proper medications decreased the CKD3 risk. Member B had 104 medical encounters in the past 3 years and was predicted to have a low risk of CKD3, because the model didn’t identify diagnoses, procedures, or medications related to kidney disease, and many lab test results, including ‘Glomerular filtration rate’ were within the normal range. The model accurately predicts members A and B and provides interpretable clinical evidence that is validated by clinicians. Without extra effort, the interpretation is generated directly from the model and presented together with the occurrence date. Our model uses the medical data in its most raw format without any further data aggregation, transformation, or mapping. This greatly simplifies the data preparation process, mitigates the chance for error and eliminates post-modeling work needed for traditional model explanation. To our knowledge, this is the first paper on an interpretable deep-learning model using a 3-level attention structure, sourcing both EMR and claim data, including all 4 types of medical data, on the entire Medicare population of a big insurance company, and more importantly, directly generating model interpretation to support user decision. In the future, we plan to enrich the model input by adding patients’ demographics and information from free-texted physician notes.

Keywords: deep learning, interpretability, attention, big data, medical conditions

Procedia PDF Downloads 75

Authors: R. Burahee, B. Chassinat, T. de Laclos, A. Dépée, A. Sastim

Abstract:

The purpose of this paper is to propose a solution to detect, predict and reconstitute rain drops in real time – during the night – using an embedded material with an infrared camera. To prevent the system from needing too high hardware resources, simple models are considered in a powerful image treatment algorithm reducing considerably calculation time in OpenCV software. Using a smart model – drops will be matched thanks to a process running through two consecutive pictures for implementing a sophisticated tracking system. With this system drops computed trajectory gives information for predicting their future location. Thanks to this technique, treatment part can be reduced. The hardware system composed by a Raspberry Pi is optimized to host efficiently this code for real time execution.

Keywords: reconstitution, prediction, detection, rain drop, real time, raspberry, infrared

Procedia PDF Downloads 391

Authors: Madhawa Basnayaka, Jouni Paltakari

Abstract:

Fully passive backscattering chipless RFID tags are an emerging wireless technology with low cost, higher reading distance, and fast automatic identification without human interference, unlike already available technologies like optical barcodes. The design optimization of chipless RFID tags is crucial as it requires replacing integrated chips found in conventional RFID tags with printed geometric designs. These designs enable data encoding and decoding through backscattered electromagnetic (EM) signatures. The applications of chipless RFID tags have been limited due to the constraints of data encoding capacity and the ability to design accurate yet efficient configurations. The traditional approach to accomplishing design parameters for a desired EM response involves iterative adjustment of design parameters and simulating until the desired EM spectrum is achieved. However, traditional numerical simulation methods encounter limitations in optimizing design parameters efficiently due to the speed and resource consumption. In this work, a deep learning neural network (DNN) is utilized to establish a correlation between the EM spectrum and the dimensional parameters of nested centric rings, specifically square and octagonal. The proposed bi-directional DNN has two simultaneously running neural networks, namely spectrum prediction and design parameters prediction. First, spectrum prediction DNN was trained to minimize mean square error (MSE). After the training process was completed, the spectrum prediction DNN was able to accurately predict the EM spectrum according to the input design parameters within a few seconds. Then, the trained spectrum prediction DNN was connected to the design parameters prediction DNN and trained two networks simultaneously. For the first time in chipless tag design, design parameters were predicted accurately after training bi-directional DNN for a desired EM spectrum. The model was evaluated using a randomly generated spectrum and the tag was manufactured using the predicted geometrical parameters. The manufactured tags were successfully tested in the laboratory. The amount of iterative computer simulations has been significantly decreased by this approach. Therefore, highly efficient but ultrafast bi-directional DNN models allow rapid and complicated chipless RFID tag designs.

Keywords: artificial intelligence, chipless RFID, deep learning, machine learning

Procedia PDF Downloads 23

Authors: Khalid Ali, Manar Jammal

Abstract:

In order to meet the stringent latency and reliability requirements of the upcoming 5G networks, Open Radio Access Networks (O-RAN) have been proposed. The virtualization of O-RAN has allowed it to be treated as a Network Function Virtualization (NFV) architecture, while its components are considered Virtualized Network Functions (VNFs). Hence, intelligent Machine Learning (ML) based solutions can be utilized to apply different resource management and allocation techniques on O-RAN. However, intelligently allocating resources for O-RAN VNFs can prove challenging due to the dynamicity of traffic in mobile networks. Network providers need to dynamically scale the allocated resources in response to the incoming traffic. Elastically allocating resources can provide a higher level of flexibility in the network in addition to reducing the OPerational EXpenditure (OPEX) and increasing the resources utilization. Most of the existing elastic solutions are reactive in nature, despite the fact that proactive approaches are more agile since they scale instances ahead of time by predicting the incoming traffic. In this work, we propose and evaluate traffic forecasting models based on the ML algorithm. The algorithms aim at predicting future O-RAN traffic by using previous traffic data. Detailed analysis of the traffic data was carried out to validate the quality and applicability of the traffic dataset. Hence, two ML models were proposed and evaluated based on their prediction capabilities.

Keywords: O-RAN, traffic forecasting, NFV, ARIMA, LSTM, elasticity

Procedia PDF Downloads 187

Authors: Intan Maisarah Abd Rahim, Herlina Abdul Rahim, Rashidah Ghazali

Abstract:

Diabetes is a medical condition that can lead to various diseases such as stroke, heart disease, blindness and obesity. In clinical practice, the concern of the diabetic patients towards the blood glucose examination is rather alarming as some of the individual describing it as something painful with pinprick and pinch. As for some patient with high level of glucose level, pricking the fingers multiple times a day with the conventional glucose meter for close monitoring can be tiresome, time consuming and painful. With these concerns, several non-invasive techniques were used by researchers in measuring the glucose level in blood, including ultrasonic sensor implementation, multisensory systems, absorbance of transmittance, bio-impedance, voltage intensity, and thermography. This paper is discussing the application of the near-infrared (NIR) spectroscopy as a non-invasive method in measuring the glucose level and the implementation of the linear system identification model in predicting the output data for the NIR measurement. In this study, the wavelengths considered are at the 1450 nm and 1950 nm. Both of these wavelengths showed the most reliable information on the glucose presence in blood. Then, the linear Autoregressive Moving Average Exogenous model (ARMAX) model with both un-regularized and regularized methods was implemented in predicting the output result for the NIR measurement in order to investigate the practicality of the linear system in this study. However, the result showed only 50.11% accuracy obtained from the system which is far from the satisfying results that should be obtained.

Keywords: diabetes, glucose level, linear, near-infrared, non-invasive, prediction system

Procedia PDF Downloads 136

Authors: I. Behroyan, P. Ganesan, S. He, S. Sivasankaran

Abstract:

This study compares the predictions of five types of Computational Fluid Dynamics (CFD) models, including two single-phase models (i.e. Newtonian and non-Newtonian) and three two-phase models (Eulerian-Eulerian, mixture and Eulerian-Lagrangian), to investigate turbulent forced convection of Cu-water nanofluid in a tube with a constant heat flux on the tube wall. The Reynolds (Re) number of the flow is between 10,000 and 25,000, while the volume fraction of Cu particles used is in the range of 0 to 2%. The commercial CFD package of ANSYS-Fluent is used. The results from the CFD models are compared with results from experimental investigations from literature. According to the results of this study, non-Newtonian single-phase model, in general, does not show a good agreement with Xuan and Li correlation in prediction of Nu number. Eulerian-Eulerian model gives inaccurate results expect for φ=0.5%. Mixture model gives a maximum error of 15%. Newtonian single-phase model and Eulerian-Lagrangian model, in overall, are the recommended models. This work can be used as a reference for selecting an appreciate model for future investigation. The study also gives a proper insight about the important factors such as Brownian motion, fluid behavior parameters and effective nanoparticle conductivity which should be considered or changed by the each model.

Keywords: heat transfer, nanofluid, single-phase models, two-phase models

Procedia PDF Downloads 466