Reduce Misclassification Rates Research Articles

Applying advanced sentiment analysis for strategic marketing insights: A case study of BBVA using machine learning techniques

In the digital era, understanding public sentiment toward brands on social media is essential for crafting effective marketing strategies. This study applies sentiment analysis on Banco Bilbao Vizcaya Argentaria (BBVA) tweets using advanced machine learning techniques, particularly the eXtreme Gradient Boosting (XGBoost) algorithm, which showed remarkable precision (91.2%) in sentiment classification. This process involved a systematic approach to data collection, cleaning, and preprocessing. The precision of XGBoost highlights its effectiveness in analyzing social media conversations about banking. Additionally, this paper achieved improvements in neutral tweet classification, with accuracy rates at 87-88% and a reduced misclassification rate, enhancing the analysis reliability. The findings not only uncover general sentiments toward BBVA but also provide insight into how these sentiments shift in response to marketing activities and global events. This gives marketers a valuable tool for real-time assessment of campaign effectiveness and brand perception. Ultimately, employing the XGBoost algorithm for sentiment analysis offers BBVA a strategic advantage in understanding and engaging its online audience, demonstrating the significant benefits of using sophisticated machine learning in banking. The study emphasizes the crucial role of data-driven sentiment analysis in developing informed business strategies and improving customer relationships in the banking industry’s competitive landscape.

Malaria detection using custom Semantic Segmentation Neural Network Architecture

Malaria is a significant disease that affects both animals and humans. The four main Plasmodium species that cause human malaria are Plasmodium falciparum, Plasmodium vivax, Plasmodium malariae, and Plasmodium ovale. Plasmodium knowlesi, a parasite typically infecting forest macaque monkeys, was recently revealed to be able to be transmitted by anophelines and provoke malaria in humans. This provides an increasing risk of spreading the disease to areas previously unaffected with it and infecting people during the increasingly popular travels abroad. Microscopic examination remains one of the most often used methods for its laboratory confirmation. These tests, however, should be performed immediately after receiving samples from a firstcontact doctor to allow immediate therapy. This research presents a novel, semantic segmentation neural network architecture designed to quickly create a classification mask, giving the doctor information about the position, shape, and possible affiliation of detected elements. The evaluation method is based on a light microscope imagery and was created to overcome problems resulting from the human diagnosis specifics. There are 3 abstract classes containing healthy cells, cells with malaria and background. The outputted mask can be later mapped to a more readable form with the inclusion of contrasting colors, next to an original image for quick validation. Such an approach allows for semi-automatic recognition of possible disease, nevertheless still giving the final verdict to the specialist. The developed solution has achieved a high recognition accuracy of 96.65%, while the computer power requirements are kept at a minimum. The proposed solution can help reduce misclassification rates by providing additional data for the doctor and speed up the entire process with the early diagnosis made by a deep learning model.

Security Assessment Algorithm of Navigation Control System Based on Big Data

There is a trend of integration and modularisation in navigation control system. It has the characteristics of high resource sharing, fast information transmission and integration of software and hardware. It has high requirements for the safety of navigation control systems. In view of the low accuracy of traditional algorithms, a new big data-based security assessment algorithm of navigation control system is proposed. The structure of navigation control system is introduced and the realisation of navigation control system is analysed. In big data environment, common criteria (CCS) is used to determine the safety assessment objectives of the navigation control system. According to the different safety weights among the areas in the criterion layer, the pilots determine the weight of the assessment objectives by analytic hierarchy process based on the evaluation of the navigation control system and the corresponding safety rating indices. GRAP algorithm is used to acquire the security level of navigation control system so as to realise the safety assessment of the navigation control system. Experimental results show that the proposed algorithm can effectively improve the accuracy of assessment and reduce misclassification rate due to the comprehensive ability of this algorithm.

Human Activity Recognition based on Interaction Modelling

Human activity recognition aims at recognising and interpreting the activities of humans automatically from videos. Among the activities of humans, identifying the interactions between human within minimal computation time and reduced misclassification rate is a cumbersome task. Hence, an interaction-based human activity recognition system is proposed in this paper that utilises silhouette features to identify and classify the interactions between humans. The main issues that affect the performance of activity recognition are sudden illumination changes, detection of static human, data discrimination, data variance, crowding problem, and computational complexity. To accomplish the preceding issues, three new algorithms named weight-based updating Gaussian mixture model (wu-GMM), spatial dissemination-based contour silhouettes (SDCS), and weighted constrained dynamic time warping (WCDTW) are proposed. Experiments are conducted with the gaming dataset and Kinect interaction dataset to show that the proposed system recognises the interactions with reduced misclassification rate and minimal processing time compared to the existing system.

Security assessment algorithm of navigation control system based on big data

There is a trend of integration and modularisation in navigation control system. It has the characteristics of high resource sharing, fast information transmission and integration of software and hardware. It has high requirements for the safety of navigation control systems. In view of the low accuracy of traditional algorithms, a new big data-based security assessment algorithm of navigation control system is proposed. The structure of navigation control system is introduced and the realisation of navigation control system is analysed. In big data environment, common criteria (CCS) is used to determine the safety assessment objectives of the navigation control system. According to the different safety weights among the areas in the criterion layer, the pilots determine the weight of the assessment objectives by analytic hierarchy process based on the evaluation of the navigation control system and the corresponding safety rating indices. GRAP algorithm is used to acquire the security level of navigation control system so as to realise the safety assessment of the navigation control system. Experimental results show that the proposed algorithm can effectively improve the accuracy of assessment and reduce misclassification rate due to the comprehensive ability of this algorithm.

Analysis of Time-Series Spectral Index Data to Enhance Crop Identification Over a Mediterranean Rural Landscape

Spectral index time series can provide valuable phenological information into the classification process for the precise crop mapping, in order to reduce misclassification rates associated with low interclass and high intraclass spectral variability. Stochastic hidden Markov models (HMMs) are efficient yet computationally demanding classification approach which can simulate crop dynamics, exploiting the spectral information of their phenological states and the relations between these states. This letter aims to present a methodology which achieves accurate classification results while maintaining a low computational cost. A classification framework based on HMMs was developed, and different spectral indices were generated from the time series of Landsat ETM+ and RapidEye imagery, for modeling crop vegetation dynamics over a Mediterranean rural area, with high spatiotemporal crop heterogeneity. To further improve the HMMs indices classification, separability analysis and two different decision fusion strategies were tested. The assessment of the classification accuracy, along with an evaluation of the computational cost, indicated that the green-red vegetation index produced the most favorable results among the individual spectral indices. Although the decision fusion based on an integration of a reliability factor increased the overall accuracy by 3.1%, this came at the cost of computational time, compared to the separability analysis model which required less processing time.

Ridge-forward quadratic discriminant analysis in high-dimensional situations

Quadratic discriminant analysis is a classical and popular classification tool, but it fails to work in high-dimensional situations where the dimension p is larger than the sample size n. To address this issue, the authors propose a ridge-forward quadratic discriminant (RFQD) analysis method via screening relevant predictors in a successive manner to reduce misclassification rate. The authors use extended Bayesian information criterion to determine the final model and prove that RFQD is selection consistent. Monte Carlo simulations are conducted to examine its performance.

Improved CT-based estimate of pulmonary gas trapping accounting for scanner and lung-volume variations in a multicenter asthmatic study.

Lung air trapping is estimated via quantitative computed tomography (CT) using density threshold-based measures on an expiration scan. However, the effects of scanner differences and imaging protocol adherence on quantitative assessment are known to be problematic. This study investigates the effects of protocol differences, such as using different CT scanners and breath-hold coaches in a multicenter asthmatic study, and proposes new methods that can adjust intersite and intersubject variations. CT images of 50 healthy subjects and 42 nonsevere and 52 severe asthmatics at total lung capacity (TLC) and functional residual capacity (FRC) were acquired using three different scanners and two different coaching methods at three institutions. A fraction threshold-based approach based on the corrected Hounsfield unit of air with tracheal density was applied to quantify air trapping at FRC. The new air-trapping method was enhanced by adding a lung-shaped metric at TLC and the lobar ratio of air-volume change between TLC and FRC. The fraction-based air-trapping method is able to collapse air-trapping data of respective populations into distinct regression lines. Relative to a constant value-based clustering scheme, the slope-based clustering scheme shows the improved performance and reduced misclassification rate of healthy subjects. Furthermore, both lung shape and air-volume change are found to be discriminant variables for differentiating among three populations of healthy subjects and nonsevere and severe asthmatics. In conjunction with the lung shape and air-volume change, the fraction-based measure of air trapping enables differentiation of severe asthmatics from nonsevere asthmatics and nonsevere asthmatics from healthy subjects, critical for the development and evaluation of new therapeutic interventions.

Boosting Techniques for Physics‐Based Vortex Detection

AbstractRobust automated vortex detection algorithms are needed to facilitate the exploration of large‐scale turbulent fluid flow simulations. Unfortunately, robust non‐local vortex detection algorithms are computationally intractable for large data sets and local algorithms, while computationally tractable, lack robustness. We argue that the deficiencies inherent to the local definitions occur because of two fundamental issues: the lack of a rigorous definition of a vortex and the fact that a vortex is an intrinsically non‐local phenomenon. As a first step towards addressing this problem, we demonstrate the use of machine learning techniques to enhance the robustness of local vortex detection algorithms. We motivate the presence of an expert‐in‐the‐loop using empirical results based on machine learning techniques. We employ adaptive boosting to combine a suite of widely used, local vortex detection algorithms, which we term weak classifiers, into a robust compound classifier. Fundamentally, the training phase of the algorithm, in which an expert manually labels small, spatially contiguous regions of the data, incorporates non‐local information into the resulting compound classifier. We demonstrate the efficacy of our approach by applying the compound classifier to two data sets obtained from computational fluid dynamical simulations. Our results demonstrate that the compound classifier has a reduced misclassification rate relative to the component classifiers.

An Improved ID3 Algorithm

Desion tree,which is used to classify samples, is one of the important models in data mining. As the core algorithm of decision tree,the classical ID3 algorithm is being widely used in classification problems by its simplicity and efficiency. Unfortunately, it is prone to making the attribute which contains more values as decision attribute, so the attribute which has strong classification ability are probably missed. Proposes an improved ID3 algorithm. When the most information gains of the attributions are same, the algorithm helps us to select an attribute which can get better classification effect. Compared to the classical ID3 algorithm, the new one can reduce misclassification rate and simplify the complexity of the decision tree.

Optimal feature selection for sparse linear discriminant analysis and its applications in gene expression data

This work studies the theoretical rules of feature selection in linear discriminant analysis (LDA), and a new feature selection method is proposed for sparse linear discriminant analysis. An l1 minimization method is used to select the important features from which the LDA will be constructed. The asymptotic results of this proposed two-stage LDA (TLDA) are studied, demonstrating that TLDA is an optimal classification rule whose convergence rate is the best compared to existing methods. The experiments on simulated and real datasets are consistent with the theoretical results and show that TLDA performs favorably in comparison with current methods. Overall, TLDA uses a lower minimum number of features or genes than other approaches to achieve a better result with a reduced misclassification rate.

A Road to Classification in High Dimensional Space: The Regularized Optimal Affine Discriminant

SummaryFor high dimensional classification, it is well known that naively performing the Fisher discriminant rule leads to poor results due to diverging spectra and accumulation of noise. Therefore, researchers proposed independence rules to circumvent the diverging spectra, and sparse independence rules to mitigate the issue of accumulation of noise. However, in biological applications, often a group of correlated genes are responsible for clinical outcomes, and the use of the covariance information can significantly reduce misclassification rates. In theory the extent of such error rate reductions is unveiled by comparing the misclassification rates of the Fisher discriminant rule and the independence rule. To materialize the gain on the basis of finite samples, a regularized optimal affine discriminant (ROAD) is proposed. The ROAD selects an increasing number of features as the regularization relaxes. Further benefits can be achieved when a screening method is employed to narrow the feature pool before applying the ROAD method. An efficient constrained co-ordinate descent algorithm is also developed to solve the associated optimization problems. Sampling properties of oracle type are established. Simulation studies and real data analysis support our theoretical results and demonstrate the advantages of the new classification procedure under a variety of correlation structures. A delicate result on continuous piecewise linear solution paths for the ROAD optimization problem at the population level justifies the linear interpolation of the constrained co-ordinate descent algorithm.

E0399 The ACE gene polymorphisms distribution status in Hainan Li and Han peoples with coronary heart disease

ObjectiveTo explore the significance of the ACE gene insertion/deletion (I/D) polymorphism in peoples of Hainan Li and Han nationality with coronary heart disease (CHD).MethodsUsed the PCR to detect the polymorphisms...

Detection of dust aerosol by combining CALIPSO active lidar and passive IIR measurements

Abstract. The version 2 Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) dust layer detection method, which is based only on lidar measurements, misclassified about 43% dust layers (mainly dense dust layers) as cloud layers over the Taklamakan Desert. To address this problem, a new method was developed by combining the CALIPSO Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) and passive Infrared Imaging Radiometer (IIR) measurements. This combined lidar and IR measurement (hereafter, CLIM) method uses the IIR tri-spectral IR brightness temperatures to discriminate between ice cloud and dense dust layers, and lidar measurements alone to detect thin dust and water cloud layers. The brightness temperature difference between 10.60 and 12.05 μm (BTD11−12) is typically negative for dense dust and generally positive for ice cloud, but it varies from negative to positive for thin dust layers, which the CALIPSO lidar correctly identifies. Results show that the CLIM method could significantly reduce misclassification rates to as low as ~7% for the active dust season of spring 2008 over the Taklamakan Desert. The CLIM method also revealed 18% more dust layers having greatly intensified backscatter between 1.8 and 4 km altitude over the source region compared to the CALIPSO version 2 data. These results allow a more accurate assessment of the effect of dust on climate.

Simulation-Based Secondary Incident Filtering Method

A significant portion of nonrecurrent freeway congestion is caused by incidents. This nonrecurrent congestion negatively impacts safety and mobility. It produces enormous travel delay and results in secondary incidents, which cause additional delay and injury. The simulation-based secondary incident filtering (SBSIF) method is proposed for identifying secondary incidents from archived incident data. The proposed methodology is computationally efficient and overcomes deficiencies of existing techniques. It is evaluated on incident data from a 6-month period along a segment of I-287 in the New York state. Results of the evaluation show that the SBSIF method has a significantly reduced misclassification rate (e.g., a reduction of 58 percentage points and greater) as compared with static methods commonly used in practice, despite that, it requires comparable computational effort.

Morphological, Physiological, and Genetic Techniques for Improving Field Identification of Steelhead, Coastal Cutthroat Trout, and Hybrid Smolts

AbstractIn streams with sympatric populations of steelhead Oncorhynchus mykiss and coastal cutthroat trout O. clarkii clarkii (hereafter, cutthroat trout), life history descriptions and smolt production estimates may be hampered by misclassification of hybrids as steelhead or cutthroat trout. Additionally, important morphological and physiological differences between hybrid and non‐hybird smolts are often unknown. Therefore, we assessed field classification and created classification models to quantify and reduce misclassification rates among migrating steelhead, cutthroat trout, and hybrid smolts. Field misclassifications of smolts with steelhead or cutthroat trout genotypes were low (1% and 2%, respectively). However, field misclassification of fish with hybrid genotypes was high, with 11% of the hybrids being misclassified as steelhead and 42% of the hybrids being misclassified as cutthroat trout. Hybrid smolts were larger, had lower gill Na+, K+‐ATPase activities, and lower condition factors than steelhead but were similar to cutthroat trout smolts in these same measurements. Additionally, statistical classification analyses using morphological traits including subterminal jaw slash intensity, hyoid teeth presence, maxillary length, breaks of pigment along outer margin of adipose fin, condition factor, and migration date improved classification error rates of hybrids from 53% to 21%. In systems with sympatric populations of steelhead and cutthroat trout, we recommend a thorough evaluation of field‐based identification methods with genetic techniques to assess the effectiveness of field‐based classification in addition to examining important life history differences among steelhead, cutthroat trout, and their hybrids.

Standardization of spirometry in assessment of responders following man‐made disasters: World Trade Center worker and volunteer medical screening program

Spirometry is the most commonly used pulmonary function test to screen individuals for suspected lung disease. It is also used for screening workers with exposures to agents associated with pulmonary diseases. Although the American Thoracic Society (ATS) provides guidelines for spirometers and spirometry techniques, many factors are not standardized, so that results from individual pulmonary function laboratories vary substantially. These differences can create substantial difficulties in using data pooled from multiple sites to understand health consequences of disasters that involve exposures to pulmonary toxins. This article describes the approach used to minimize these differences for a consortium of institutions who are providing medical monitoring examinations to World Trade Center (WTC) responders. The protocol improved upon the minimal ATS guidelines. Spirometric measurements were obtained before and after use of a bronchodilator. A fourth-generation spirometer was chosen that exceeded ATS spirometer accuracy standards. The accuracy was verified at the beginning of each day of testing. Technologists who performed the spirometry tests were centrally trained and certified and received regular reports on their performance. Reference values and normal ranges were obtained from the National Health and Nutrition Examination Survey (NHANES III) data set. A standardized interpretation flowchart was followed to reduce misclassification rates for airway obstruction and restriction. Patients with spirometric abnormalities were referred for more extensive diagnostic testing. More than 12,000 spirometry tests were performed during the first examination. The 20 spirometers used at the 6 participating institutions maintained accuracy within 3% for more than 4 years. Overall, more than 80% of the test sessions met ATS quality goals. Spirometry abnormality rates exceeded those obtained for adults who participated in the NHANES III survey. The program allowed standardization of the performance and interpretation of spirometry results across multiple institutions. This facilitated reliable and rapid diagnosis of lung disease in the large number of WTC responders screened. We recommend this approach for postdisaster pulmonary evaluations in other settings.

Determining positivity of alcohol abuse by Taguchi methods

Aims to establish the critical score and screening accuracy of the CAGE Questionnaire in three treatment settings--primary health care, walk-in (triage) clinic and the emergency room. Taguchi methods are applied to three screens of the CAGE questionnaire. Analysis of the sensitivity and specificity data of three CAGE screens by leveling factor (p'), signal-to-noise ratios (S/N, SS/N) and their dependent relation resulted in critical CAGE scores of 1, 1 and 2; and high screening accuracy levels of 98.44, 97.20 and 94.92 percent, respectively. The illustrated method yielded excellent (> or = 95 percent) screening accuracy values for primary health care, emergency room and walk-in clinic patients. To reduce misclassification rates of alcohol abuse, screening systems should concentrate first on developing ways to standardize protocols. Further work is needed to establish high screening accuracy in other clinical settings, and particularly in those at risk of alcohol abuse in the general population.

3-D underwater object recognition

In this paper, we propose an automatic supervised classification of objects lying on the sea floor or buried in sediment layers. This pattern recognition provides a way to distinguish natural and manufactured objects and then should be helpful to detect mine, pipe-line, or wreckage. Proposed methods combine different techniques: pattern information extraction, relevant parameter search, and supervised classifier. Parameters are automatically selected using a principal component analysis to reduce misclassification rate and to simplify classifier structure. Performances of different parameters (two-dimensional and three-dimensional) are compared and discussed from testing on synthetic and real data bases.

Importance of replication in microarray gene expression studies: statistical methods and evidence from repetitive cDNA hybridizations.

We present statistical methods for analyzing replicated cDNA microarray expression data and report the results of a controlled experiment. The study was conducted to investigate inherent variability in gene expression data and the extent to which replication in an experiment produces more consistent and reliable findings. We introduce a statistical model to describe the probability that mRNA is contained in the target sample tissue, converted to probe, and ultimately detected on the slide. We also introduce a method to analyze the combined data from all replicates. Of the 288 genes considered in this controlled experiment, 32 would be expected to produce strong hybridization signals because of the known presence of repetitive sequences within them. Results based on individual replicates, however, show that there are 55, 36, and 58 highly expressed genes in replicates 1, 2, and 3, respectively. On the other hand, an analysis by using the combined data from all 3 replicates reveals that only 2 of the 288 genes are incorrectly classified as expressed. Our experiment shows that any single microarray output is subject to substantial variability. By pooling data from replicates, we can provide a more reliable analysis of gene expression data. Therefore, we conclude that designing experiments with replications will greatly reduce misclassification rates. We recommend that at least three replicates be used in designing experiments by using cDNA microarrays, particularly when gene expression data from single specimens are being analyzed.