Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 18

Search results for: interpretability

18 Use of Interpretable Evolved Search Query Classifiers for Sinhala Documents

Authors: Prasanna Haddela

Abstract:

Document analysis is a well matured yet still active research field, partly as a result of the intricate nature of building computational tools but also due to the inherent problems arising from the variety and complexity of human languages. Breaking down language barriers is vital in enabling access to a number of recent technologies. This paper investigates the application of document classification methods to new Sinhalese datasets. This language is geographically isolated and rich with many of its own unique features. We will examine the interpretability of the classification models with a particular focus on the use of evolved Lucene search queries generated using a Genetic Algorithm (GA) as a method of document classification. We will compare the accuracy and interpretability of these search queries with other popular classifiers. The results are promising and are roughly in line with previous work on English language datasets.

Keywords: evolved search queries, Sinhala document classification, Lucene Sinhala analyzer, interpretable text classification, genetic algorithm

Procedia PDF Downloads 46
17 MhAGCN: Multi-Head Attention Graph Convolutional Network for Web Services Classification

Authors: Bing Li, Zhi Li, Yilong Yang

Abstract:

Web classification can promote the quality of service discovery and management in the service repository. It is widely used to locate developers desired services. Although traditional classification methods based on supervised learning models can achieve classification tasks, developers need to manually mark web services, and the quality of these tags may not be enough to establish an accurate classifier for service classification. With the doubling of the number of web services, the manual tagging method has become unrealistic. In recent years, the attention mechanism has made remarkable progress in the field of deep learning, and its huge potential has been fully demonstrated in various fields. This paper designs a multi-head attention graph convolutional network (MHAGCN) service classification method, which can assign different weights to the neighborhood nodes without complicated matrix operations or relying on understanding the entire graph structure. The framework combines the advantages of the attention mechanism and graph convolutional neural network. It can classify web services through automatic feature extraction. The comprehensive experimental results on a real dataset not only show the superior performance of the proposed model over the existing models but also demonstrate its potentially good interpretability for graph analysis.

Keywords: attention mechanism, graph convolutional network, interpretability, service classification, service discovery

Procedia PDF Downloads 68
16 Performance Evaluation of a Very High-Resolution Satellite Telescope

Authors: Walid A. Attia, Taher M. Bazan, Fawzy Eltohamy, Mahmoud Fathy

Abstract:

System performance evaluation is an essential stage in the design of high-resolution satellite telescopes prior to the development process. In this paper, a system performance evaluation of a very high-resolution satellite telescope is investigated. The evaluated system has a Korsch optical scheme design. This design has been discussed in another paper with respect to three-mirror anastigmat (TMA) scheme design and the former configuration showed better results. The investigated system is based on the Korsch optical design integrated with a time-delay and integration charge coupled device (TDI-CCD) sensor to achieve a ground sampling distance (GSD) of 25 cm. The key performance metrics considered are the spatial resolution, the signal to noise ratio (SNR) and the total modulation transfer function (MTF) of the system. In addition, the national image interpretability rating scale (NIIRS) metric is assessed to predict the image quality according to the modified general image quality equation (GIQE). Based on the orbital, optical and detector parameters, the estimated GSD is found to be 25 cm. The SNR has been analyzed at different illumination conditions of target albedos, sun and sensor angles. The system MTF has been computed including diffraction, aberration, optical manufacturing, smear and detector sampling as the main contributors for evaluation the MTF. Finally, the system performance evaluation results show that the computed MTF value is found to be around 0.08 at the Nyquist frequency, the SNR value was found to be 130 at albedo 0.2 with a nadir viewing angles and the predicted NIIRS is in the order of 6.5 which implies a very good system image quality.

Keywords: modulation transfer function, national image interpretability rating scale, signal to noise ratio, satellite telescope performance evaluation

Procedia PDF Downloads 161
15 Item-Trait Pattern Recognition of Replenished Items in Multidimensional Computerized Adaptive Testing

Authors: Jianan Sun, Ziwen Ye

Abstract:

Multidimensional computerized adaptive testing (MCAT) is a popular research topic in psychometrics. It is important for practitioners to clearly know the item-trait patterns of administered items when a test like MCAT is operated. Item-trait pattern recognition refers to detecting which latent traits in a psychological test are measured by each of the specified items. If the item-trait patterns of the replenished items in MCAT item pool are well detected, the interpretability of the items can be improved, which can further promote the abilities of the examinees who attending the MCAT to be accurately estimated. This research explores to solve the item-trait pattern recognition problem of the replenished items in MCAT item pool from the perspective of statistical variable selection. The popular multidimensional item response theory model, multidimensional two-parameter logistic model, is assumed to fit the response data of MCAT. The proposed method uses the least absolute shrinkage and selection operator (LASSO) to detect item-trait patterns of replenished items based on the essential information of item responses and ability estimates of examinees collected from a designed MCAT procedure. Several advantages of the proposed method are outlined. First, the proposed method does not strictly depend on the relative order between the replenished items and the selected operational items, so it allows the replenished items to be mixed into the operational items in reasonable order such as considering content constraints or other test requirements. Second, the LASSO used in this research improves the interpretability of the multidimensional replenished items in MCAT. Third, the proposed method can exert the advantage of shrinkage method idea for variable selection, so it can help to check item quality and key dimension features of replenished items and saves more costs of time and labors in response data collection than traditional factor analysis method. Moreover, the proposed method makes sure the dimensions of replenished items are recognized to be consistent with the dimensions of operational items in MCAT item pool. Simulation studies are conducted to investigate the performance of the proposed method under different conditions for varying dimensionality of item pool, latent trait correlation, item discrimination, test lengths and item selection criteria in MCAT. Results show that the proposed method can accurately detect the item-trait patterns of the replenished items in the two-dimensional and the three-dimensional item pool. Selecting enough operational items from the item pool consisting of high discriminating items by Bayesian A-optimality in MCAT can improve the recognition accuracy of item-trait patterns of replenished items for the proposed method. The pattern recognition accuracy for the conditions with correlated traits is better than those with independent traits especially for the item pool consisting of comparatively low discriminating items. To sum up, the proposed data-driven method based on the LASSO can accurately and efficiently detect the item-trait patterns of replenished items in MCAT.

Keywords: item-trait pattern recognition, least absolute shrinkage and selection operator, multidimensional computerized adaptive testing, variable selection

Procedia PDF Downloads 55
14 A Time-Varying and Non-Stationary Convolution Spectral Mixture Kernel for Gaussian Process

Authors: Kai Chen, Shuguang Cui, Feng Yin

Abstract:

Gaussian process (GP) with spectral mixture (SM) kernel demonstrates flexible non-parametric Bayesian learning ability in modeling unknown function. In this work a novel time-varying and non-stationary convolution spectral mixture (TN-CSM) kernel with a significant enhancing of interpretability by using process convolution is introduced. A way decomposing the SM component into an auto-convolution of base SM component and parameterizing it to be input dependent is outlined. Smoothly, performing a convolution between two base SM component yields a novel structure of non-stationary SM component with much better generalized expression and interpretation. The TN-CSM perfectly allows compatibility with the stationary SM kernel in terms of kernel form and spectral base ignored and confused by previous non-stationary kernels. On synthetic and real-world datatsets, experiments show the time-varying characteristics of hyper-parameters in TN-CSM and compare the learning performance of TN-CSM with popular and representative non-stationary GP.

Keywords: Gaussian process, spectral mixture, non-stationary, convolution

Procedia PDF Downloads 75
13 Software Assessment Using Ant Colony Optimization Algorithm

Authors: Saad M. Darwish

Abstract:

Recently, software quality issues have come to be seen as important subject as we see an enormous growth of agencies involved in software industries. However,these agencies cannot guarantee the quality of their products, thus leaving users in uncertainties. Software certification is the extension of quality by means that quality needs to be measured prior to certification granting process. This research participates in solving the problem of software assessment by proposing a model for assessment and certification of software product that uses a fuzzy inference engine to integrate both of process–driven and application-driven quality assurance strategies. The key idea of the on hand model is to improve the compactness and the interpretability of the model’s fuzzy rules via employing an ant colony optimization algorithm (ACO), which tries to find good rules description by dint of compound rules initially expressed with traditional single rules. The model has been tested by case study and the results have demonstrated feasibility and practicability of the model in a real environment.

Keywords: optimization technique, quality assurance, software certification model, software assessment

Procedia PDF Downloads 408
12 Evolving Software Assessment and Certification Models Using Ant Colony Optimization Algorithm

Authors: Saad M. Darwish

Abstract:

Recently, software quality issues have come to be seen as important subject as we see an enormous growth of agencies involved in software industries. However, these agencies cannot guarantee the quality of their products, thus leaving users in uncertainties. Software certification is the extension of quality by means that quality needs to be measured prior to certification granting process. This research participates in solving the problem of software assessment by proposing a model for assessment and certification of software product that uses a fuzzy inference engine to integrate both of process–driven and application-driven quality assurance strategies. The key idea of the on hand model is to improve the compactness and the interpretability of the model’s fuzzy rules via employing an ant colony optimization algorithm (ACO), which tries to find good rules description by dint of compound rules initially expressed with traditional single rules. The model has been tested by case study and the results have demonstrated feasibility and practicability of the model in a real environment.

Keywords: software quality, quality assurance, software certification model, software assessment

Procedia PDF Downloads 432
11 Deep Learning Based Unsupervised Sport Scene Recognition and Highlights Generation

Authors: Ksenia Meshkova

Abstract:

With increasing amount of multimedia data, it is very important to automate and speed up the process of obtaining meta. This process means not just recognition of some object or its movement, but recognition of the entire scene versus separate frames and having timeline segmentation as a final result. Labeling datasets is time consuming, besides, attributing characteristics to particular scenes is clearly difficult due to their nature. In this article, we will consider autoencoders application to unsupervised scene recognition and clusterization based on interpretable features. Further, we will focus on particular types of auto encoders that relevant to our study. We will take a look at the specificity of deep learning related to information theory and rate-distortion theory and describe the solutions empowering poor interpretability of deep learning in media content processing. As a conclusion, we will present the results of the work of custom framework, based on autoencoders, capable of scene recognition as was deeply studied above, with highlights generation resulted out of this recognition. We will not describe in detail the mathematical description of neural networks work but will clarify the necessary concepts and pay attention to important nuances.

Keywords: neural networks, computer vision, representation learning, autoencoders

Procedia PDF Downloads 55
10 A Selection Approach: Discriminative Model for Nominal Attributes-Based Distance Measures

Authors: Fang Gong

Abstract:

Distance measures are an indispensable part of many instance-based learning (IBL) and machine learning (ML) algorithms. The value difference metrics (VDM) and inverted specific-class distance measure (ISCDM) are among the top-performing distance measures that address nominal attributes. VDM performs well in some domains owing to its simplicity and poorly in others that exist missing value and non-class attribute noise. ISCDM, however, typically works better than VDM on such domains. To maximize their advantages and avoid disadvantages, in this paper, a selection approach: a discriminative model for nominal attributes-based distance measures is proposed. More concretely, VDM and ISCDM are built independently on a training dataset at the training stage, and the most credible one is recorded for each training instance. At the test stage, its nearest neighbor for each test instance is primarily found by any of VDM and ISCDM and then chooses the most reliable model of its nearest neighbor to predict its class label. It is simply denoted as a discriminative distance measure (DDM). Experiments are conducted on the 34 University of California at Irvine (UCI) machine learning repository datasets, and it shows DDM retains the interpretability and simplicity of VDM and ISCDM but significantly outperforms the original VDM and ISCDM and other state-of-the-art competitors in terms of accuracy.

Keywords: distance measure, discriminative model, nominal attributes, nearest neighbor

Procedia PDF Downloads 50
9 A Tool to Measure Efficiency and Trust Towards eXplainable Artificial Intelligence in Conflict Detection Tasks

Authors: Raphael Tuor, Denis Lalanne

Abstract:

The ATM research community is missing suitable tools to design, test, and validate new UI prototypes. Important stakes underline the implementation of both DSS and XAI methods into current systems. ML-based DSS are gaining in relevance as ATFM becomes increasingly complex. However, these systems only prove useful if a human can understand them, and thus new XAI methods are needed. The human-machine dyad should work as a team and should understand each other. We present xSky, a configurable benchmark tool that allows us to compare different versions of an ATC interface in conflict detection tasks. Our main contributions to the ATC research community are (1) a conflict detection task simulator (xSky) that allows to test the applicability of visual prototypes on scenarios of varying difficulty and outputting relevant operational metrics (2) a theoretical approach to the explanations of AI-driven trajectory predictions. xSky addresses several issues that were identified within available research tools. Researchers can configure the dimensions affecting scenario difficulty with a simple CSV file. Both the content and appearance of the XAI elements can be customized in a few steps. As a proof-of-concept, we implemented an XAI prototype inspired by the maritime field.

Keywords: air traffic control, air traffic simulation, conflict detection, explainable artificial intelligence, explainability, human-automation collaboration, human factors, information visualization, interpretability, trajectory prediction

Procedia PDF Downloads 69
8 Detection of Atrial Fibrillation Using Wearables via Attentional Two-Stream Heterogeneous Networks

Authors: Huawei Bai, Jianguo Yao, Fellow, IEEE

Abstract:

Atrial fibrillation (AF) is the most common form of heart arrhythmia and is closely associated with mortality and morbidity in heart failure, stroke, and coronary artery disease. The development of single spot optical sensors enables widespread photoplethysmography (PPG) screening, especially for AF, since it represents a more convenient and noninvasive approach. To our knowledge, most existing studies based on public and unbalanced datasets can barely handle the multiple noises sources in the real world and, also, lack interpretability. In this paper, we construct a large- scale PPG dataset using measurements collected from PPG wrist- watch devices worn by volunteers and propose an attention-based two-stream heterogeneous neural network (TSHNN). The first stream is a hybrid neural network consisting of a three-layer one-dimensional convolutional neural network (1D-CNN) and two-layer attention- based bidirectional long short-term memory (Bi-LSTM) network to learn representations from temporally sampled signals. The second stream extracts latent representations from the PPG time-frequency spectrogram using a five-layer CNN. The outputs from both streams are fed into a fusion layer for the outcome. Visualization of the attention weights learned demonstrates the effectiveness of the attention mechanism against noise. The experimental results show that the TSHNN outperforms all the competitive baseline approaches and with 98.09% accuracy, achieves state-of-the-art performance.

Keywords: PPG wearables, atrial fibrillation, feature fusion, attention mechanism, hyber network

Procedia PDF Downloads 48
7 Transfer Learning in Actuarial Science: Primer and Applications

Authors: Youngsun Kim

Abstract:

With the increase in data availability, the use of machine learning methods has gained popularity in both the academia and business world. Many studies from different disciplines are using machine learning methods to address their question of interest. But expanding the usage of such models based on traditional machine learning methods requires a large amount of labeled data, which is costly from an economic standpoint. This is also relevant to the actuarial science field. With the limited amount of labeled data, using advanced learners—particularly deep neural networks—that hasled to major advances in other domains is difficult. The small margin of predictive superiority, particularly when considered with other problems such as model stability and interpretability such “black-box” models face, explains why sofar, deep learning is not the method of choice in actuarial modeling. Researchers from other fields propose transfer learning as a potential solution in similar contexts. Transfer learning is taking the knowledge gained from a previous problem and applying it to a new but related problem. By transferring knowledge at the data level as well as the model level, transfer learning can reduce the cost of collecting a large amount of new labeled data and increase the predictive performance of a model. In this paper, various transfer learning approaches are introduced and applied to publicly available insurance datasets. The performance of each approach will be evaluated compared to a baseline model in the actuarial context. This will demonstrate the possibility of transfer learning as another solution to the limitation of machine learning approaches that are present in the realm of actuarial science.

Keywords: machine learning, transfer learning, actuarial modeling, predictive performance

Procedia PDF Downloads 26
6 Human-Machine Cooperation in Facial Comparison Based on Likelihood Scores

Authors: Lanchi Xie, Zhihui Li, Zhigang Li, Guiqiang Wang, Lei Xu, Yuwen Yan

Abstract:

Image-based facial features can be classified into category recognition features and individual recognition features. Current automated face recognition systems extract a specific feature vector of different dimensions from a facial image according to their pre-trained neural network. However, to improve the efficiency of parameter calculation, an algorithm generally reduces the image details by pooling. The operation will overlook the details concerned much by forensic experts. In our experiment, we adopted a variety of face recognition algorithms based on deep learning, compared a large number of naturally collected face images with the known data of the same person's frontal ID photos. Downscaling and manual handling were performed on the testing images. The results supported that the facial recognition algorithms based on deep learning detected structural and morphological information and rarely focused on specific markers such as stains and moles. Overall performance, distribution of genuine scores and impostor scores, and likelihood ratios were tested to evaluate the accuracy of biometric systems and forensic experts. Experiments showed that the biometric systems were skilled in distinguishing category features, and forensic experts were better at discovering the individual features of human faces. In the proposed approach, a fusion was performed at the score level. At the specified false accept rate, the framework achieved a lower false reject rate. This paper contributes to improving the interpretability of the objective method of facial comparison and provides a novel method for human-machine collaboration in this field.

Keywords: likelihood ratio, automated facial recognition, facial comparison, biometrics

Procedia PDF Downloads 56
5 Variational Explanation Generator: Generating Explanation for Natural Language Inference Using Variational Auto-Encoder

Authors: Zhen Cheng, Xinyu Dai, Shujian Huang, Jiajun Chen

Abstract:

Recently, explanatory natural language inference has attracted much attention for the interpretability of logic relationship prediction, which is also known as explanation generation for Natural Language Inference (NLI). Existing explanation generators based on discriminative Encoder-Decoder architecture have achieved noticeable results. However, we find that these discriminative generators usually generate explanations with correct evidence but incorrect logic semantic. It is due to that logic information is implicitly encoded in the premise-hypothesis pairs and difficult to model. Actually, logic information identically exists between premise-hypothesis pair and explanation. And it is easy to extract logic information that is explicitly contained in the target explanation. Hence we assume that there exists a latent space of logic information while generating explanations. Specifically, we propose a generative model called Variational Explanation Generator (VariationalEG) with a latent variable to model this space. Training with the guide of explicit logic information in target explanations, latent variable in VariationalEG could capture the implicit logic information in premise-hypothesis pairs effectively. Additionally, to tackle the problem of posterior collapse while training VariaztionalEG, we propose a simple yet effective approach called Logic Supervision on the latent variable to force it to encode logic information. Experiments on explanation generation benchmark—explanation-Stanford Natural Language Inference (e-SNLI) demonstrate that the proposed VariationalEG achieves significant improvement compared to previous studies and yields a state-of-the-art result. Furthermore, we perform the analysis of generated explanations to demonstrate the effect of the latent variable.

Keywords: natural language inference, explanation generation, variational auto-encoder, generative model

Procedia PDF Downloads 67
4 Comparison of Multivariate Adaptive Regression Splines and Random Forest Regression in Predicting Forced Expiratory Volume in One Second

Authors: P. V. Pramila , V. Mahesh

Abstract:

Pulmonary Function Tests are important non-invasive diagnostic tests to assess respiratory impairments and provides quantifiable measures of lung function. Spirometry is the most frequently used measure of lung function and plays an essential role in the diagnosis and management of pulmonary diseases. However, the test requires considerable patient effort and cooperation, markedly related to the age of patients esulting in incomplete data sets. This paper presents, a nonlinear model built using Multivariate adaptive regression splines and Random forest regression model to predict the missing spirometric features. Random forest based feature selection is used to enhance both the generalization capability and the model interpretability. In the present study, flow-volume data are recorded for N= 198 subjects. The ranked order of feature importance index calculated by the random forests model shows that the spirometric features FVC, FEF 25, PEF,FEF 25-75, FEF50, and the demographic parameter height are the important descriptors. A comparison of performance assessment of both models prove that, the prediction ability of MARS with the `top two ranked features namely the FVC and FEF 25 is higher, yielding a model fit of R2= 0.96 and R2= 0.99 for normal and abnormal subjects. The Root Mean Square Error analysis of the RF model and the MARS model also shows that the latter is capable of predicting the missing values of FEV1 with a notably lower error value of 0.0191 (normal subjects) and 0.0106 (abnormal subjects). It is concluded that combining feature selection with a prediction model provides a minimum subset of predominant features to train the model, yielding better prediction performance. This analysis can assist clinicians with a intelligence support system in the medical diagnosis and improvement of clinical care.

Keywords: FEV, multivariate adaptive regression splines pulmonary function test, random forest

Procedia PDF Downloads 241
3 Deciphering Orangutan Drawing Behavior Using Artificial Intelligence

Authors: Benjamin Beltzung, Marie Pelé, Julien P. Renoult, Cédric Sueur

Abstract:

To this day, it is not known if drawing is specifically human behavior or if this behavior finds its origins in ancestor species. An interesting window to enlighten this question is to analyze the drawing behavior in genetically close to human species, such as non-human primate species. A good candidate for this approach is the orangutan, who shares 97% of our genes and exhibits multiple human-like behaviors. Focusing on figurative aspects may not be suitable for orangutans’ drawings, which may appear as scribbles but may have meaning. A manual feature selection would lead to an anthropocentric bias, as the features selected by humans may not match with those relevant for orangutans. In the present study, we used deep learning to analyze the drawings of a female orangutan named Molly († in 2011), who has produced 1,299 drawings in her last five years as part of a behavioral enrichment program at the Tama Zoo in Japan. We investigate multiple ways to decipher Molly’s drawings. First, we demonstrate the existence of differences between seasons by training a deep learning model to classify Molly’s drawings according to the seasons. Then, to understand and interpret these seasonal differences, we analyze how the information spreads within the network, from shallow to deep layers, where early layers encode simple local features and deep layers encode more complex and global information. More precisely, we investigate the impact of feature complexity on classification accuracy through features extraction fed to a Support Vector Machine. Last, we leverage style transfer to dissociate features associated with drawing style from those describing the representational content and analyze the relative importance of these two types of features in explaining seasonal variation. Content features were relevant for the classification, showing the presence of meaning in these non-figurative drawings and the ability of deep learning to decipher these differences. The style of the drawings was also relevant, as style features encoded enough information to have a classification better than random. The accuracy of style features was higher for deeper layers, demonstrating and highlighting the variation of style between seasons in Molly’s drawings. Through this study, we demonstrate how deep learning can help at finding meanings in non-figurative drawings and interpret these differences.

Keywords: cognition, deep learning, drawing behavior, interpretability

Procedia PDF Downloads 4
2 Ground Motion Modeling Using the Least Absolute Shrinkage and Selection Operator

Authors: Yildiz Stella Dak, Jale Tezcan

Abstract:

Ground motion models that relate a strong motion parameter of interest to a set of predictive seismological variables describing the earthquake source, the propagation path of the seismic wave, and the local site conditions constitute a critical component of seismic hazard analyses. When a sufficient number of strong motion records are available, ground motion relations are developed using statistical analysis of the recorded ground motion data. In regions lacking a sufficient number of recordings, a synthetic database is developed using stochastic, theoretical or hybrid approaches. Regardless of the manner the database was developed, ground motion relations are developed using regression analysis. Development of a ground motion relation is a challenging process which inevitably requires the modeler to make subjective decisions regarding the inclusion criteria of the recordings, the functional form of the model and the set of seismological variables to be included in the model. Because these decisions are critically important to the validity and the applicability of the model, there is a continuous interest on procedures that will facilitate the development of ground motion models. This paper proposes the use of the Least Absolute Shrinkage and Selection Operator (LASSO) in selecting the set predictive seismological variables to be used in developing a ground motion relation. The LASSO can be described as a penalized regression technique with a built-in capability of variable selection. Similar to the ridge regression, the LASSO is based on the idea of shrinking the regression coefficients to reduce the variance of the model. Unlike ridge regression, where the coefficients are shrunk but never set equal to zero, the LASSO sets some of the coefficients exactly to zero, effectively performing variable selection. Given a set of candidate input variables and the output variable of interest, LASSO allows ranking the input variables in terms of their relative importance, thereby facilitating the selection of the set of variables to be included in the model. Because the risk of overfitting increases as the ratio of the number of predictors to the number of recordings increases, selection of a compact set of variables is important in cases where a small number of recordings are available. In addition, identification of a small set of variables can improve the interpretability of the resulting model, especially when there is a large number of candidate predictors. A practical application of the proposed approach is presented, using more than 600 recordings from the National Geospatial-Intelligence Agency (NGA) database, where the effect of a set of seismological predictors on the 5% damped maximum direction spectral acceleration is investigated. The set of candidate predictors considered are Magnitude, Rrup, Vs30. Using LASSO, the relative importance of the candidate predictors has been ranked. Regression models with increasing levels of complexity were constructed using one, two, three, and four best predictors, and the models’ ability to explain the observed variance in the target variable have been compared. The bias-variance trade-off in the context of model selection is discussed.

Keywords: ground motion modeling, least absolute shrinkage and selection operator, penalized regression, variable selection

Procedia PDF Downloads 254
1 On the Influence of Sleep Habits for Predicting Preterm Births: A Machine Learning Approach

Authors: C. Fernandez-Plaza, I. Abad, E. Diaz, I. Diaz

Abstract:

Births occurring before the 37th week of gestation are considered preterm births. A threat of preterm is defined as the beginning of regular uterine contractions, dilation and cervical effacement between 23 and 36 gestation weeks. To author's best knowledge, the factors that determine the beginning of the birth are not completely defined yet. In particular, the incidence of sleep habits on preterm births is weekly studied. The aim of this study is to develop a model to predict the factors affecting premature delivery on pregnancy, based on the above potential risk factors, including those derived from sleep habits and light exposure at night (introduced as 12 variables obtained by a telephone survey using two questionnaires previously used by other authors). Thus, three groups of variables were included in the study (maternal, fetal and sleep habits). The study was approved by Research Ethics Committee of the Principado of Asturias (Spain). An observational, retrospective and descriptive study was performed with 481 births between January 1, 2015 and May 10, 2016 in the University Central Hospital of Asturias (Spain). A statistical analysis using SPSS was carried out to compare qualitative and quantitative variables between preterm and term delivery. Chi-square test qualitative variable and t-test for quantitative variables were applied. Statistically significant differences (p < 0.05) between preterm vs. term births were found for primiparity, multi-parity, kind of conception, place of residence or premature rupture of membranes and interruption during nights. In addition to the statistical analysis, machine learning methods to look for a prediction model were tested. In particular, tree based models were applied as the trade-off between performance and interpretability is especially suitable for this study. C5.0, recursive partitioning, random forest and tree bag models were analysed using caret R-package. Cross validation with 10-folds and parameter tuning to optimize the methods were applied. In addition, different noise reduction methods were applied to the initial data using NoiseFiltersR package. The best performance was obtained by C5.0 method with Accuracy 0.91, Sensitivity 0.93, Specificity 0.89 and Precision 0.91. Some well known preterm birth factors were identified: Cervix Dilation, maternal BMI, Premature rupture of membranes or nuchal translucency analysis in the first trimester. The model also identifies other new factors related to sleep habits such as light through window, bedtime on working days, usage of electronic devices before sleeping from Mondays to Fridays or change of sleeping habits reflected in the number of hours, in the depth of sleep or in the lighting of the room. IF dilation < = 2.95 AND usage of electronic devices before sleeping from Mondays to Friday = YES and change of sleeping habits = YES, then preterm is one of the predicting rules obtained by C5.0. In this work a model for predicting preterm births is developed. It is based on machine learning together with noise reduction techniques. The method maximizing the performance is the one selected. This model shows the influence of variables related to sleep habits in preterm prediction.

Keywords: machine learning, noise reduction, preterm birth, sleep habit

Procedia PDF Downloads 80