Low Accuracy Rate Research Articles

Target Tracking Method for Transmission Line Moving Operation Based on Inspection Robot and Edge Computing

Aiming at the problems of low accuracy and high loss rate when the traditional target tracking (TT) method is applied to the TT of the moving operation of the transmission line, a transmission line based on an inspection robot and edge computing (EC) is proposed: the mobile job TT method. First, the basic framework of the TT algorithm is proposed, relying on the edge device to develop the TT system for mobile operations on the transmission line. Video information is collected by an intelligent inspection robot and sent to the target tracking system in the edge device for processing to obtain accurate data. Then, the gradient disappearance and explosion problems caused by the increase of network depth are solved by using the deep residual network. The traditional deep residual network is improved by introducing the improved bidirectional feature reinforcement network and the classification and regression subnet. The loss of position texture information is remedied, and the accurate tracking of the moving target on transmission line is realized. Finally, the real-time data acquisition of mobile operation target is realized by using an intelligent inspection robot, and the experimental verification is conducted. The proposed algorithm is compared and analyzed against the three other algorithms using the same data set through simulation experiments. The results show the precision, recall rate, accuracy, and comprehensive evaluation index F1 value of the proposed algorithm rank highest, reaching 93.8%, 90.2%, 83.8%, and 89.8%, respectively, compared with the other algorithms.

Transformer DGA Fault Diagnosis Method Based on Stacked Sparse Auto-Encoders

In the field of transformer fault diagnosis, the imbalance of fault samples seriously affects the fault identification performance of the diagnosis model. Focusing on the problems of low accuracy and high leakage rate of diagnosis model caused by unbalanced fault samples of transformer, a transformer fault diagnosis method is proposed. First of all, the Borderline SMOTE algorithm is used to balance the fault data set from a few samples on the boundary, so as to achieve the effect of power transformer fault sample equalization. Secondly, a diagnosis model of the stacked sparse auto-Encoders with the gas in oil as input characteristic parameter is built. Finally, an evaluation system consisting of accuracy rate, recall rate and F1 score is selected to compare the diagnosis effect of the model before and after category equalization, the experiment proves the effectiveness of this method.

When and where did it happen? Systematic differences in recall of core and optional sentence information.

This article reports on four experiments aiming to examine immediate post-sentential recall of core sentence information (conveyed by direct objects), and optional/additional information (conveyed by temporal or locative adjuncts). Participants read simple and unambiguous Czech sentences such as Starší důchodce velmi pečlivě pročetl noviny v neděli v knihovně: "An older retiree read the newspaper very carefully on Sunday in the library." Sentences always appeared as a whole after pressing a space bar. Immediately after the sentence disappeared, an open-ended (free response) question was presented targeting either the direct object (e.g., newspaper), temporal adjunct (e.g., on Sunday), or locative adjunct (e.g., in the library). Altogether, it was found that the core information (conveyed by the direct object) was recalled almost perfectly, whereas additional information, conveyed by temporal and locative adjuncts, was recalled with significantly lower accuracy rates. Information structure also played a role: if the temporal or locative adjunct was focused, it was recalled better than if it was unfocused. The present article thus shows systematic differences in recall success for different pieces of information. These findings suggest the presence of selective attention mechanisms during early stages of sentence processing. Factors such as syntactic function or information structure influence the degree of attention to different pieces of information conveyed by a sentence. In turn, certain pieces of information may not be consciously accessible already after the sentence is processed.

A Network Intrusion Detection Method Based on Improved Bi-LSTM in Internet of Things Environment

When performing malicious network attack detection, traditional intrusion detection methods show their disadvantage of low accuracy and high false detection rate. To address these problems, this paper proposes a novel network intrusion detection scheme based on an improved bi-directional long short-term memory (Bi-LSTM) model under the emerging internet of things (IoT) environment. Firstly, this paper analyzes Bi-LSTM model. Then, it introduces a two-layer attention network structure into Bi-LSTM network. Finally, the corresponding network intrusion detection system is constructed based on the improved Bi STM model. Through simulation experiments, the proposed network intrusion detection method and other three methods are compared under five identical databases. Experimental results show that the false detection rate and detection accuracy of the proposed method are optimal on all sample data, the detection accuracy reaches 97.24% and the false detection rate drops to 5.13%.

Multimodal sentiment system and method based on CRNN-SVM

Traditional sentiment analysis focuses on text-level sentiment mining, transforming sentiment mining into classification or regression problems, resulting in a sentiment analysis low accuracy rate. Sentiment analysis refers to the use of natural language processing, text analysis, and computational linguistics to systematically identify, extract, quantify, and study sentimental states. Therefore, more scholars have begun to focus on speech recognition and facial expression recognition research, and extracting and analysing people’s sentiment tendencies can improve sentiment recognition accuracy. Traditional single-modal sentiment analysis can no longer meet people’s needs. Therefore, this paper proposes a multimodal sentiment analysis method based on the multimodal sentiment analysis method that can obtain more sentimental information sources and help people make better decisions. The experimental results in this paper show that the highest recognition rates of CNN-SVM, RNN-SVM, and CRNN-SVM were 76.8%, 71.2%, and 93.5%, respectively. It can be seen that CRNN-SVM has the highest sentiment tendency recognition rate in deep learning, so it is suitable to apply CRNN-SVM to sentiment tendency analysis system design in this paper. The average accuracy rate of the system designed in this paper was 91%, and the stability was also very strong, which shows that the system designed in this paper is meaningful. The main contribution of this paper is based on the limitations of single-mode emotion analysis. It proposes a multimode emotion analysis method and introduces a convolutional neural network to help people obtain more emotional information sources to meet their needs.

Can we predict orthodontic extraction patterns by using machine learning?

To investigate the utility of machine learning (ML) in accurately predicting orthodontic extraction patterns in a heterogeneous population. The material of this retrospective study consisted of records of 366 patients treated with orthodontic extractions. The dataset was randomly split into training (70%) and test sets (30%) and was stratified according to race/ethnicity and gender. Fifty-five cephalometric and demographic input data were used to train and test multiple ML algorithms. The extraction patterns were labelled according to the previous treatment plan. Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM) algorithms were used to predict the patient's extraction patterns. The highest class accuracy percentages were obtained for the upper and lower 1st premolars (U/L4s) (RF: 81.63%, LR: 63.27%, SVM: 63.27%) and upper 1st premolars only (U4s) extraction patterns (RF: 61.11%, LR: 72.22%, SVM: 72.22%). However, all methods revealed low class accuracy rates (<50%) for the upper 1st and lower 2nd premolars (U4/L5s), upper 2nd and lower 1st premolars (U5/L4s), and upper and lower 2nd premolars (U/L5s) extraction patterns. For the overall accuracy, RF yielded the highest percentage with 54.55%, followed by SVM with 52.73% and LR with 49.09%. All tested supervised ML techniques yielded good accuracy in predicting U/L4s and U4s extraction patterns. However, they predicted poorly for the U4/L5s, U5/L4s, and U/L5s extraction patterns. Molar relationship, mandibular crowding, and overjet were found to be the most predictive indicators for determining extraction patterns.

Bitcoin Return Prediction based on OLS, Random Forest, LightGBM, and LSTM

Since Bitcoin was proposed in 2008, it has become a very valuable asset and an important part of many investors’ portfolios. It’s important to both understand Bitcoin mechanics and predict its valuation with the help of the state-of-art machine learning tools. The study develops four different models, including Ordinary Least Squares (OLS) regression model, Random Forest, Light Gradient Boosting Machine (LightGBM), and Long Short-Term Memory (LSTM), to predict the return of Bitcoin and compare the performance of these models. According to the analysis, the daily changes in the high, low, close price of Bitcoin, and close price of Tesla stock, and gold price between yesterday and today are all strongly correlated to the Bitcoin return on tomorrow. The statistical approach, or OLS modeling, has the simplest algorithm whereas the highest accuracy rate. The LightGBM model and LSTM model have lower accuracy rates in order, but still exceed the 50% (random benchmark). The Random Forest model, as another type of decision tree algorithm, has similar prediction results with the LightGBM model but a lower accuracy rate that fails to reach the benchmark. Based on the analysis, multiple factors affect the Bitcoin return, and these results provide an insight for investors to the cryptocurrency market and the macroeconomic environment. It validates the effectiveness of several machine learning algorithms in Bitcoin return forecasting and supports future developments in related fields.

Predicting Bankruptcy Using Altman's Z-Score And Zmijewski Models

Oil and natural gas (oil and gas) are strategic natural resources and have an important role in national development. The need for oil and gas is expected to continue to increase in line with economic and population growth. Therefore, oil and gas business activities must be managed as well as possible to meet current and future supplies. Financial crises (financial distress) can occur in various companies and can be a signal of bankruptcy that the company may experience. The purpose of this study is to determine the bankruptcy prediction of Sub-Oil and Gas companies in Indonesia that are listed on the IDX in 2017 – 2020 using Altman's Z – score and Zmijewski models. The Altman Z-Score in predicting bankruptcy has a low accuracy rate of 22.72% with an error of 56.82%, while companies that are in a critical condition or gray area are not included in the calculation of the accuracy or type of error, due to the gray area cannot be determined whether the company is bankrupt or healthy, in other terms, namely the gray area. The Zmijewski in predicting bankruptcy is more accurate than the Altman Z-Score, which is 63.63% with a type error of 36.37%.

A Study of Breast Cancer Classification Algorithms by Fusing Machine Learning and Deep Learning

Although breast cancer, with easy recurrence and high mortality, has become one of the leading causes of cancer death in women, early and accurate diagnosis of breast cancer can effectively increase the likelihood of a cure. Therefore, it is particularly important to improve the accuracy of early diagnosis of breast cancer. However, conventional early diagnosis relies on human experience and has a low accuracy rate. Therefore, many researchers have proposed various machine learning methods to improve the accuracy and efficiency of prediction. Most of the existing studies around breast cancer classification adopt a single algorithm to fit breast cancer data but ignore the applicability of different breast cancer data features to the model. In this paper, we adopt machine algorithms to strip the features of machine learning methods from the rest of the features and attempt to enhance the model effect by designing deep learning model structures to find the hidden patterns in the rest of the features. In addition, due to strict medical data privacy requirements and high collection difficulty and cost, the model designed in this paper will be trained on a small number of samples. As a result, we attempt to find a minimization model for a breast cancer classification algorithm that features both low cost and high efficiency. At the same time, the deep learning model is further designed to complement the original model when it is possible to introduce complex data indicators. Experimental values show that the design model in this paper performs best not only under limited data and limited indicators but also under limited data complex indicators, demonstrating the effectiveness of the approach of mixed comparison and feature selection of multiple classification algorithms. In summary, the fusion model designed and implemented in this paper performs well in the experiments, and the accuracy of the model test reaches 98.3%.

ERP and Behavioural Correlates of Prospective Memory in Bilinguals during L1 and L2 Processing.

Language influences how we process information from multiple domains. Thus, working in first (L1) or second language (L2) can modulate bilinguals' performance on basic activities, such as visual search, decision-making, or reading. However, few studies have explored the role of L1 and L2 processing during an essential ability, such as Prospective Memory (PM). This type of memory allows us to set intentions to perform in the future (e.g., to attend an appointment). Thus, this is a novel study that allows us to explore the influence of bilingual language processing on certain cognitive abilities, which have not been deeply studied yet, such as the recall of future intentions. Thereby, this study aimed to explore the neural and behavioural correlates of bilinguals during L1 and L2 processing in a PM task where participants had to carry out an ongoing task while recovering a prospective intention given a PM cue. Importantly, the nature of the PM cue (focal or non-focal) varied the monitoring demands of the task. Behavioural and Event-Related Potential (ERP) results indicated greater engagement of monitoring processes in the PM task during L2 processing. Specifically, in L2, we found lower accuracy rates in the ongoing task and smaller amplitude differences between the focal and non-focal conditions in the P3b. Altogether, these findings suggest an impairment in prospective processing due to working in L2 contexts, supporting previous research on the impact of the bilingual experience over PM.

Privacy, bias and the clinical use of facial recognition technology: A survey of genetics professionals.

Facial recognition technology (FRT) has been adopted as a precision medicine tool. The medical genetics field highlights both the clinical potential and privacy risks of this technology, putting the discipline at the forefront of a new digital privacy debate. Investigating how geneticists perceive the privacy concerns surrounding FRT can help shape the evolution and regulation of the field, and provide lessons for medicine and research more broadly. Five hundred and sixty-two genetics clinicians and researchers were approached to fill out a survey, 105 responded, and 80% of these completed. The survey consisted of 48 questions covering demographics, relationship to new technologies, views on privacy, views on FRT, and views on regulation. Genetics professionals generally placed a high value on privacy, although specific views differed, were context-specific, and covaried with demographic factors. Most respondents (88%) agreed that privacy is a basic human right, but only 37% placed greater weight on it than other values such as freedom of speech. Most respondents (80%) supported FRT use in genetics, but not necessarily for broader clinical use. A sizeable percentage (39%) were unaware of FRT's lower accuracy rates in marginalized communities and of the mental health effects of privacy violations (62%), but most (76% and 75%, respectively) expressed concern when informed. Overall, women and those who self-identified as politically progressive were more concerned about the lower accuracy rates in marginalized groups (88% vs. 64% and 83% vs. 63%, respectively). Younger geneticists were more wary than older geneticists about using FRT in genetics (28% compared to 56% "strongly" supported such use). There was an overall preference for more regulation, but respondents had low confidence in governments' or technology companies' ability to accomplish this. Privacy views are nuanced and context-dependent. Support for privacy was high but not absolute, and clear deficits existed in awareness of crucial FRT-related discrimination potential and mental health impacts. Education and professional guidelines may help to evolve views and practices within the field.

A vector conversion method of building indoor space data based on attribute classification

Aiming at the problems of low data conversion efficiency, low accuracy and low data utilization rate after conversion in traditional methods, this paper proposes a vector conversion method for building indoor space data based on attribute classification. Firstly, the transformation process of data vectors is analyzed. Secondly, block matching detection and fusion recognition were carried out on the building interior space images, and fuzzy feature extraction method was used to optimize the collection and feature recognition of the building interior space data. Then, the attribute classification method is used to obtain the condition attribute and decision attribute of the data, and realize the building interior space data mining. Then, the K-means algorithm is used to cluster the indoor spatial data samples, and the wavelet transform method is used to de-noise the noisy data in advance. Finally, the obtained data is processed by vector transformation. The experimental results show that the data conversion efficiency of this method is high, and the data conversion accuracy and data utilization rate have been improved.

Triage of in vivo burn injuries and prediction of wound healing outcome using neural networks and modeling of the terahertz permittivity based on the double Debye dielectric parameters.

The initial assessment of the depth of a burn injury during triage forms the basis for determination of the course of the clinical treatment plan. However, severe skin burns are highly dynamic and hard to predict. This results in a low accuracy rate of about 60 - 75% in the diagnosis of partial-thickness burns in the acute post-burn period. Terahertz time-domain spectroscopy (THz-TDS) has demonstrated a significant potential for non-invasive and timely estimation of the burn severity. Here, we describe a methodology for the measurement and numerical modeling of the dielectric permittivity of the in vivo porcine skin burns. We use the double Debye dielectric relaxation theory to model the permittivity of the burned tissue. We further investigate the origins of dielectric contrast between the burns of various severity, as determined histologically based on the percentage of the burned dermis, using the empirical Debye parameters. We demonstrate that the five parameters of the double Debye model can form an artificial neural network classification algorithm capable of automatic diagnosis of the severity of the burn injuries, and predicting its ultimate wound healing outcome by forecasting its re-epithelialization status in 28 days. Our results demonstrate that the Debye dielectric parameters provide a physics-based approach for the extraction of the biomedical diagnostic markers from the broadband THz pulses. This method can significantly boost dimensionality reduction of THz training data in artificial intelligence models and streamline machine learning algorithms.

Human Action Recognition for Dynamic Scenes of Emergency Rescue Based on Spatial-Temporal Fusion Network

Targeting the problems of the insufficient utilization of temporal and spatial information in videos and a lower accuracy rate, this paper proposes a human action recognition method for dynamic videos of emergency rescue based on a spatial-temporal fusion network. A time domain segmentation strategy based on random sampling maintains the overall time domain structure of the video. Considering the spatial-temporal asynchronous relationship, multiple asynchronous motion sequences are increased as input of the temporal convolutional network. spatial-temporal features are fused in convolutional layers to reduce feature loss. Because time series information is crucial for human action recognition, the acquired mid-layer spatial-temporal fusion features are sent into Bidirectional Long Short-Term Memory (Bi-LSTM) to obtain the human movement features in the whole video temporal dimension. Experiment results show the proposed method fully fuses spatial and temporal dimension information and improves the accuracy of human action recognition in dynamic scenes. It is also faster than traditional methods.

A Video Sequence Face Expression Recognition Method Based on Squeeze-and-Excitation and 3DPCA Network.

Expression recognition is a very important direction for computers to understand human emotions and human-computer interaction. However, for 3D data such as video sequences, the complex structure of traditional convolutional neural networks, which stretch the input 3D data into vectors, not only leads to a dimensional explosion, but also fails to retain structural information in 3D space, simultaneously leading to an increase in computational cost and a lower accuracy rate of expression recognition. This paper proposes a video sequence face expression recognition method based on Squeeze-and-Excitation and 3DPCA Network (SE-3DPCANet). The introduction of a 3DPCA algorithm in the convolution layer directly constructs tensor convolution kernels to extract the dynamic expression features of video sequences from the spatial and temporal dimensions, without weighting the convolution kernels of adjacent frames by shared weights. Squeeze-and-Excitation Network is introduced in the feature encoding layer, to automatically learn the weights of local channel features in the tensor features, thus increasing the representation capability of the model and further improving recognition accuracy. The proposed method is validated on three video face expression datasets. Comparisons were made with other common expression recognition methods, achieving higher recognition rates while significantly reducing the time required for training.

Resting Energy Expenditure in the Critically Ill and Healthy Elderly—A Retrospective Matched Cohort Study

The use of indirect calorimetry to measure resting energy expenditure (mREE) is widely recommended as opposed to calculating REE (cREE) by predictive equations (PE). The aim of this study was to compare mREE with cREE in critically ill, mechanically ventilated patients aged ≥ 75 years and a healthy control group matched by age, gender and body mass index. The primary outcome was the PE accuracy rate of mREE/cREE, derived using Bland Altman plots. Secondary analyses included linear regression analyses for determinants of intraindividual mREE/cREE differences in the critically ill and interindividual mREE differences in the matched healthy cohort. In this retrospective study, 90 critically ill patients (median age 80 years) and 58 matched healthy persons were included. Median mREE was significantly higher in the critically ill (1457 kcal/d) versus the healthy cohort (1351 kcal/d), with low PE accuracy rates (21% to 49%). Independent predictors of mREE/cREE differences in the critically ill were body temperature, heart rate, FiO2, hematocrit, serum sodium and urea. Body temperature, respiratory rate, and FiO2 were independent predictors of interindividual mREE differences (critically ill versus healthy control). In conclusion, the commonly used PE in the elderly critically ill are inaccurate. Respiratory, metabolic and energy homeostasis variables may explain intraindividual mREE/cREE as well as interindividual mREE differences.

Temporal processing deficit in children and adolescents with autism spectrum disorder: An online assessment.

The sensory deficit has been considered as one of the core features in children and adolescents with autism spectrum disorder (ASD). The present study aimed to examine the temporal processing of simple and more complex auditory inputs in ASD children and adolescents with an online assessment that can be conducted remotely. One hundred fifty-eight children and adolescents aged 5-17 years participated in this study, including 79 ASD participants and 79 typically developing (TD) participants. The online assessment consisted of two temporal-order judgment tasks that required repeating the sequence of two pure tones or consonant-vowel (CV) syllabic pairs at varying interstimulus intervals (ISIs). Significantly lower accuracy rates were found in ASD than TD participants in the pure tone and the CV conditions with both short and long ISI. In addition, ASD participants (M = 245.97 ms) showed a significantly higher passing threshold than TD participants (M = 178.84 ms) in the CV task. Receiver operating characteristic analysis found that the age × ISI passing threshold composite yielded a sensitivity of 74.7% and a specificity of 50.6% at the cutoff point of -0.307 in differentiating ASD participants from TD participants. In sum, children and adolescents with ASD showed impaired temporal processing of both simple and more complex auditory stimuli, and the online assessment seems to be sensitive in differentiating individuals with ASD from those with TD.

A Lightweight Facial Emotion Recognition System Using Partial Transfer Learning for Visually Impaired People

The inability to perceive visual and other non-verbal cues for individuals with visual impairment can pose a significant challenge for their correct conversational interactions and can be an impediment for various daily life activities. Recent advancements in computational resources, particularly the computer vision capabilities can be utilized to design effective applications for visually impaired people (VIP).Among various assistive technologies, automated facial impression recognition with real-time accurate interpretation can be proven useful to tackle the above problem. Using such approach, facial emotions (e.g., sad, happy) can be robustly recognized and conveyed to the associated individuals. In this paper, a partial transfer learning approach was adopted utilizing a custom trained Convolutional Neural Network (CNN) for facial emotion recognition. A novel model that transfers features from one dataset to another is proposed. This model enables the transfer of features learned from a small number of instances to solve new challenges instances. Using the proposed approach based on a newly trained CNN, a portable lightweight facial expression recognition system with wireless connectivity and high detection accuracy was constructed and targeted specifically for VIP. The proposed recognition model provides a notable improvement over the current state-of-the-art, by providing the highest recognition accuracy of 82.1% on the enhanced Facial Expression Recognition 2013 (FER2013) FER2013 dataset. Moreover, with only 1.49M parameters, the model is operable on edge devices with limited memory and processing power. Overall, three labeled emotions happy, sad, surprise were recognized by the model with high accuracy whereas a relatively lower accuracy rate for anger, disgust, fear was noticed with higher misclassification labels for sad.

Analysis of factors influencing the determination of indications for allergen-specific immunotherapy

To develop a questionnaire and a scoring system for evaluating physicians' knowledge of allergen immunotherapy (AIT). Questionnaire was designed using the Questionnaire Star tool. A total of 1024 physicians were assessed, and based on the score divided into accurate judgment and inaccurate judgment groups. Statistical analysis was done, and counting data were expressed as frequencies and percentage values. Chi-square test and multi-factor logistic analysis were used to determine influencing factors on the indications for AIT. Physician's age, grade of the hospital, and pediatric specialty influenced the accurate judgment of AIT indication after adjustment for independent variables (P < 0.05). In all, 80.5% physicians exercised accurate assessment for allergic rhinitis. Allergic conjunctivitis was judged accurately by 47.0% physicians. Bronchial asthma was judged accurately by 71.0% physicians, and atopic dermatitis by 61.3% physicians, with a higher accuracy rate for pediatricians than nonpediatricians for all the mentioned conditions (P < 0.05). There was no significant difference in the accuracy of judgment between pediatricians and non-pediatricians in terms of AIT for food allergy and dust mite sensitization (P > 0.05). The results of our study demonstrated a high accuracy judgment rate among clinicians for rhinitis, asthma, and dermatitis, and a low accuracy rate for desensitization of healthy people with allergic conjunctivitis, food allergies, and allergen sensitization.

A Visual Target Representation using Saliency Detection Approach

The task of saliency detection is to identify the most important and informative part of a scene. Saliency detection is broadly applied to numerous vision problems, including image segmentation, object recognition, image compression, content-based image retrieval, and moving object detection. Existing saliency detection methods suffer a low accuracy rate because of missing components of saliency regions. This study proposes a visual saliency detection method for the target representation to represent targets more accurately. The proposed method consists of five modules. In the first module, the salient region is extracted through manifold ranking on a graph, which incorporates local grouping cues and boundary priors. Secondly, using a region of interest (ROI) algorithm and the subtraction of the salient region from the original image, other parts of the image, either related or nonrelated to the interested target, are segmented. Lastly, those related and non-related regions are classified and distinguished using our proposed algorithm. Experimental result shows that proposed salient region accurately represent the interested target which can be used for object detection and tracking applications.