Confusion Problems Research Articles

Emotion Recognition from Large-Scale Video Clips with Cross-Attention and Hybrid Feature Weighting Neural Networks.

The emotion of humans is an important indicator or reflection of their mental states, e.g., satisfaction or stress, and recognizing or detecting emotion from different media is essential to perform sequence analysis or for certain applications, e.g., mental health assessments, job stress level estimation, and tourist satisfaction assessments. Emotion recognition based on computer vision techniques, as an important method of detecting emotion from visual media (e.g., images or videos) of human behaviors with the use of plentiful emotional cues, has been extensively investigated because of its significant applications. However, most existing models neglect inter-feature interaction and use simple concatenation for feature fusion, failing to capture the crucial complementary gains between face and context information in video clips, which is significant in addressing the problems of emotion confusion and emotion misunderstanding. Accordingly, in this paper, to fully exploit the complementary information between face and context features, we present a novel cross-attention and hybrid feature weighting network to achieve accurate emotion recognition from large-scale video clips, and the proposed model consists of a dual-branch encoding (DBE) network, a hierarchical-attention encoding (HAE) network, and a deep fusion (DF) block. Specifically, the face and context encoding blocks in the DBE network generate the respective shallow features. After this, the HAE network uses the cross-attention (CA) block to investigate and capture the complementarity between facial expression features and their contexts via a cross-channel attention operation. The element recalibration (ER) block is introduced to revise the feature map of each channel by embedding global information. Moreover, the adaptive-attention (AA) block in the HAE network is developed to infer the optimal feature fusion weights and obtain the adaptive emotion features via a hybrid feature weighting operation. Finally, the DF block integrates these adaptive emotion features to predict an individual emotional state. Extensive experimental results of the CAER-S dataset demonstrate the effectiveness of our method, exhibiting its potential in the analysis of tourist reviews with video clips, estimation of job stress levels with visual emotional evidence, or assessments of mental healthiness with visual media.

Gender-neutral perceptions and toy purchasing behaviour: a socio-cultural transition in UK Millennials’ attitudes

This paper summarizes the arguments and counterarguments within the scientific discussion on the issue of gender-neutral parenting. The main purpose of the research is to investigate UK Millennial parents’ perceptions of gender-neutral parenting and their toy buying behaviour. Systematization of the literary sources and approaches for solving the problem of confusion between gender-neutral and gender-oriented parenting with a focus on children’s toys’ labelling, indicates that even with the subjective norm and strong initiatives of UK government who has been launching campaigns against gender-oriented parenting, still there has not agreed on the best solution for implementing gender-neutral mindset. Investigation of the topic of the research in the paper is carried out in the following logical sequence: introduction with research gap, literature review, methodology, findings and analysis, and conclusion. Methodological tools of the research were quantitative approach with SPSS ver.26. The object of research is the survey dataset collected from 1127 UK Millennial parents. The paper presents the results of an empirical analysis using Kruskal–Wallis test depending on children’s gender, parents’ birth years, and regression model. The result showed that, although there is recognition that gender-neutral parenting should be advocated as a social rule, the degree of final purchase behaviour towards gender-neutral toys is lower than these parents’ beliefs would suggest. The research empirically confirms and theoretically proves that at least UK Millennials believe in the gender-free theory, however, their real behaviour does not necessarily go abide by the perceived rule. In other words, while people may agree with the concept of gender-neutral parenting, they do not necessarily agree with gender-neutral parenting as a behavioural principle. The results of the research can be useful for developing further social recognition of gender-naturalness, and other relevant industrial marketing strategies.

Prototype-Based Intent Perception

Intent perception is a novel task that aims to understand the intention of images, regular classification methods usually perform unsatisfactorily on intent perception due to the semantic ambiguity problem,i.e. the intra-class variety problem in which images of the same intent class may contain objects of different semantic categories and the inter-class confusion problem in which images of different intent classes may contain objects of similar semantic categories. To address this problem, this paper introduces prototype learning into the intent perception and proposes a unified framework named PIP-Net to reduce the influence of semantic ambiguity. Specifically, for each intent class, we first filter semantic ambiguity samples which are far away from the cluster center. Then we use features of the filtered samples to generate prototypes via clustering algorithm. Besides, we enhance the diversity between prototypes of different classes to better handle the inter-class confusion problem. To update the prototypes in the training process, we introduce a global matching algorithm to holistically match each feature with class prototypes, and use the momentum update strategy to stably update prototypes. Experimental results on the Intentonomy dataset demonstrate that our method can consistently outperform the traditional classification paradigm in multiple baseline models, and verify the effectiveness of our proposed prototype learning paradigm in addressing the intent perception problem. Our proposed PIP-Net achieves a new state-of-the-art performance on Intentonomy, including Macro F1 score of 31.57% and averaging F1 score of 41.85%. Code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/CodeMonsterPHD/PIP-Net</uri> .

Efforts to Overcome Career Confusion for High School Students by Increasing Career Involvement through CIP Individual Career Counseling

School counselors can help high school students who have career problems with career guidance or counseling. So far, previous research has mostly proven the effectiveness of career group counseling. Meanwhile, individual counseling is rarely seen for its effectiveness, even though it is possible to do it by considering unique personal characteristics. The purpose of this research is to look at the effectiveness of individual career counseling with the CIP approach to increasing career involvement to overcome the problem of career confusion for high school students. This research is an experimental research with a single subject. Data collection was carried out using career engagement questionnaires, interviews, and observations during the counseling process. Data analysis uses visual techniques. The results show that individual counseling can solve the problem of career indecision by increasing career engagement. The implications of this research can be applied to school counselors at the high school level.

Soft Fault Diagnosis of Analog Circuit Based on EEMD and Improved MF-DFA

Aiming at the problems of nonlinearity and serious confusion of fault characteristics in analog circuits, this paper proposed a fault diagnosis method for an analog circuit based on ensemble empirical pattern decomposition (EEMD) and improved multifractal detrended fluctuations analysis (MF-DFA). This method consists of three steps: preprocessing, feature extraction, and fault classification identification. First, the EEMD decomposition preprocesses (denoises) the original signal; then, the appropriate IMF components are selected by correlation analysis; then, the IMF components are processed by the improved MF-DFA, and the fault feature values are extracted by calculating the multifractal spectrum parameters, and then the feature values are input to a support vector machine (SVM) for classification, which enables the diagnosis of soft faults in analog circuits. The experimental results show that the proposed EEMD-improved MF-DFA method effectively extracts the features of soft faults in nonlinear analog circuits and obtains a high diagnosis rate.

JCBIE: a joint continual learning neural network for biomedical information extraction

Extracting knowledge from heterogeneous data sources is fundamental for the construction of structured biomedical knowledge graphs (BKGs), where entities and relations are represented as nodes and edges in the graphs, respectively. Previous biomedical knowledge extraction methods simply considered limited entity types and relations by using a task-specific training set, which is insufficient for large-scale BKGs development and downstream task applications in different scenarios. To alleviate this issue, we propose a joint continual learning biomedical information extraction (JCBIE) network to extract entities and relations from different biomedical information datasets. By empirically studying different joint learning and continual learning strategies, the proposed JCBIE can learn and expand different types of entities and relations from different datasets. JCBIE uses two separated encoders in joint-feature extraction, hence can effectively avoid the feature confusion problem comparing with using one hard-parameter sharing encoder. Specifically, it allows us to adopt entity augmented inputs to establish the interaction between named entity recognition and relation extraction. Finally, a novel evaluation mechanism is proposed for measuring cross-corpus generalization errors, which was ignored by traditional evaluation methods. Our empirical studies show that JCBIE achieves promising performance when continual learning strategy is adopted with multiple corpora.

TSN-A: An Efficient Deep Learning Model for Automatic Modulation Classification Based on Intra-Class Confusion Reduction of Modulation Families

Automatic modulation classification (AMC) is an impressive technology, which is widely used in military and civilian fields. Recently, deep learning-based AMC (DL-AMC) methods have shown outstanding performance. However, most DL-AMC methods employ a single-stage network framework, which is effective for reducing only inter-class confusion of modulation families, while the intra-class confusion of modulation families attracts less attention and remains to be addressed. This letter proposes a two-stage network with attention (TSN-A), which consists of a main network and two sub-networks. In the first stage, the main network can effectively tackle the problem of inter-class confusion. In the second stage, the sub-networks can efficiently reduce the intra-class confusion of AM-SSB and AM-DSB families. TSN-A achieves the top results on the public challenging RML 2018.01a dataset and is more competitive in low SNR environments. Meanwhile, our main network exhibits stronger and more robust classification ability on different datasets than the benchmark models. Moreover, our sub-networks are very applicable to be migrated to other DL-AMC models for helping them improve the classification accuracy by reducing the intra-class confusion.

Hybrid Attention Fusion in Dense Crowd Counting

One of appealing approaches to guiding deep parameter optimization, is attentional supervision, which inspires intelligence in complex networks at a fraction of the cost, but there is still room for improvement. First, the real dense scene with varying scales and uneven density distribution of human heads, the density map cannot be clearly expressed. Second, the heavily occluded areas are extremely similar to the complex background, which further aggravates the counting error. Therefore, we propose a dual-track attention network that distinguishes between global and local information, which is responsible for the target overlap and background confusion problems, respectively, and finally converges and normalizes with the feature map to transform the multi-channel attention map into a single-channel density map. Meanwhile the heterogeneous pyramid design alleviates the distress of scale variation and density dissimilarity. Experiments on several official datasets prove the effectiveness of the scheme to enhance key information and overcome confounding factors.

Info-FPN: An Informative Feature Pyramid Network for object detection in remote sensing images

Feature pyramid networks are widely applied in remote sensing images for object detection to deal with the challenge of large scale variation in objects. However, the feature pyramid-based object detector for remote sensing images ignores the channel information loss, feature misalignment, and additional computational overhead to eliminate the aliasing effect, leading to inadequate feature extraction for multi-scale objects in remote sensing images. To address these challenges, an Informative Feature Pyramid Network (Info-FPN) is proposed. Specifically, we propose a Pixel Shuffle-based lateral connection Module (PSM) for the complete preservation of channel information in the feature pyramid. Then, to alleviate the problem of confusion caused by feature misalignment, a Feature Alignment Module (FAM) is proposed to achieve aligned feature fusion by template matching and learning feature offsets in the feature fusion stage. To eliminate the aliasing effect, we design a Semantic Encoder Module (SEM), which reduces the parameters and computation of model with the desirable detection accuracy. Extensive experiments on two challenging remote sensing datasets, namely DOTA and HRSC2016, prove the effectiveness of the proposed method which achieves comparable detection performance with the state-of-the-art FPN-based method.

Multiple Functional Significance of the Object: A Study in the Light of the Theory of Propping

The term “object” for grammarians denotes: 1- What the “subject” has fallen on. 2-The result of the conversion. 3- The “predicate” and the “ascribed to it” in the attribution indicating what the subject’s verb fell on. 4- Who participated in the action and other functions. Because of the multiplicity of functions of the “object”, its concept has become ambiguous among grammarians with other concepts such as “Mansob because of the deletion of the preposition”. This study seeks to find out the reason that prompted the grammarians to launch the term “object” on all these functions despite their differences and to solve the problem of confusion in the concepts of the object with other concepts, using the theory of “propping”, which is a modern ontological and epistemological theory that seeks to evaluate the foundations of science and solve its terminological problems. Keywords: Object, Multiplicity of functions, Theory of propping.

A Meta-analysis of Molecular Spectroscopy Databases, and Prospects of Molecule Detection with Some Future Facilities

Molecules reside broadly in the interstellar space and can be detected via spectroscopic observations. To date, more than 271 molecular species have been identified in interstellar medium or circumstellar envelopes. Molecular spectroscopic parameters measured in laboratory make the identification of new species and derivation of physical parameters possible. These spectroscopic parameters are systematically collected into databases, two of the most commonly used being the CDMS and JPL databases. While new spectroscopic parameters are continuously measured/calculated and added to those databases, at any point in time it is the existing spectroscopic data that ultimately limits what molecules can possibly be identified in astronomical data. In this work, we conduct a meta-analysis of the CDMS and JPL databases. We show the statistics of transition frequencies and their uncertainties in these two databases, and discuss the line confusion problem under certain physical environments. We then assess the prospects of detecting molecules in common ISM environments using a few facilities that are expected to be conducting spectroscopic observations in the future. Results show that CSST/HSTDM and SKA1-mid have the potential to detect some complex organic molecules, or even amino acids, with reasonable assumptions about ISM environments.

Tracking Some Phonetic Terms in the Arabic Lesson: A Descriptive Analytical Research

This research aims to shed light on some phonetic linguistic terms used in the Arabic phonetic lesson for the purpose of monitoring, analyzing and tracking its developments. Such a step helps to standardize and weigh between them. The study follows a descriptive-analytical approach; it surveys the problem of the phonetic linguistic term, and its linguistic exactness. Then, it examines some phonetic terms in the Arabic phonetic lesson, such as phonology and phonology; intensity, looseness and mediation; the production, articulatory, transition, position, and the two vocal chords. One of the most prominent conclusions of the study is that the phonetic linguistic terminology enjoyed a tangible development since its infancy, given that phonetics was in the process of emerging and did not reach maturity. Besides, the auditory impression and the western borrowing have a major role in the diversity of the phonetic linguistic terminologies. This is because the multiplicity issue of the linguistic term in general and the phonetic term in particular in Arabic linguistic studies has become one of the problems that Arab researchers suffer from. It has become a source of hesitation whether to choose a given phonetic term or another. This further reflects being unsettled or the lack of agreement on the standards and foundations of the use and translation of the phonetic term. More so, most of the sources on which the research relied on are translated sources that basically reflect the problems of confusion, ambiguity and inaccuracy in the meaning of the term.&#x0D;

Optical reflection essence of surface-mirror imaging

Although surface-mirror imaging is a rather old topic in optics textbooks, its optical reflection essence is still an open problem with especial significance. For example, about the chirality of plane-mirror images, some recent articles still claimed that it might be a psychological process to determine left-right or up-down symmetry between the object and its image, rather than a physical optics problem. In this paper, some experiments first show the incompleteness that the classical ray diagrams are just a kind of representationalism methods because they cannot represent the optical reflection essence of surface-mirror imaging process. Then, the optical reflection essence of surface-mirror imaging is revealed as a kind of polarization singularity of optical Mobius strip effect generated by incident light and medium interaction. Furthermore, we discover that secondary diffusion-reflection sources of all kinds of surface-mirror images exist inside glass or other media, which are formed from the chiral light partially inverted by optical Mobius strip effect. Our findings may explain reasonably or correct those historical puzzles or confused problems of surface-mirror imaging such as the chirality, half-wave loss and so on. Our theory is in well agreement with the Nobel laureate Feynman’s work on quantum interpretation of specular reflection. It is clarified based on our theory why seeing a mirror image with the secondary diffuse-reflection source is exactly the same as directly observing the corresponding object provided with the first secondary diffuse-reflecting source, without any human’s imagination. In addition, the new concept of “image barrier” is proposed and demonstrated by experiments, which determines the type of mirror images observable and the others hidden. Finally, it can be concluded that the traditional concepts of real image and virtual image are inaccurate and unnecessary, because their reflection natures are exactly the same during imaging process, and their observability is just affected by "image barrier" or other accessorial devices such as eyes and camera's convex lens, or geometric focusing properties of a concave mirror while being observed.

Time-varying voltage flicker analysis based on analytic-adaptive variational mode decomposition

Time-varying voltage flicker caused by non-linear and fluctuating loads have adverse impact on the power grid and lighting equipment. This paper proposes a time-varying voltage flicker analysis method based on analytic-adaptive variational mode decomposition (AAVMD). A novel improved square demodulation method based on analytic mode decomposition (AMD), is proposed to quickly extract the time-varying flicker envelope signals, which simplify the computational process of voltage flicker without the analog filter. Then, the adaptive variational mode decomposition (AVMD), using the energy loss coefficient and energy difference as a criterion for determining the number of modal decompositions, is proposed to detect the time-varying voltage flicker envelope which effectively avoids the pattern confusion problem of Hilbert-Huang transform (HHT). The accuracy and validity of the proposed method are demonstrated in a National Instrument-PXI system under the conditions of single frequency flicker modulation, time-varying flicker modulation magnitude, multi-frequency flicker modulation, fundamental frequency deviation and white noise interference.

Research on named entity recognition of chinese electronic medical records based on multi-head attention mechanism and character-word information fusion

In the medical field, Named Entity Recognition (NER) plays a crucial role in the process of information extraction through electronic medical records and medical texts. To address the problems of long distance entity, entity confusion, and difficulty in boundary division in the Chinese electronic medical record NER task, we propose a Chinese electronic medical record NER method based on the multi-head attention mechanism and character-word fusion. This method uses a new character-word joint feature representation based on the pre-training model BERT and self-constructed domain dictionary, which can accurately divide the entity boundary and solve the impact of unregistered words. Subsequently, on the basis of the BiLSTM-CRF model, a multi-head attention mechanism is introduced to learn the dependency relationship between remote entities and entity information in different semantic spaces, which effectively improves the performance of the model. Experiments show that our models have better performance and achieves significant improvement compared to baselines. The specific performance is that the F1 value on the Chinese electronic medical record data set reaches 95.22%, which is 2.67%higher than the F1 value of the baseline model.

A Proposed Model to eliminate the Confusion of Hematological Diseases in Thin Blood Smear by Using Deep Learning-Pretrained Model

This research aimed to developing and designing a model for resolving the confusion between hematology in a thin blood smear by means of a pre-defined deep learning model for detection and identification of hematological diseases in thin blood smear images for accurate diagnosis of the different diseases that leukemia and malaria were performed as a sample. There are catastrophic consequences that may lead to death as a result of a mistake in diagnosis and confusion in the knowledge of the disease in particular, especially in hematology, where another disease that was not originally found in the sample is identified for the similarity, which results in surgery and sometimes the administration of drugs in error. In this work, an image processing system was developed to identify patients with malaria and leukemia. The techniques in deep learning have been implemented where the CNN (Alexnet and Resnet50) image recognition model was applied to detect patterns and extract features of the different types of malaria and leukemia from the images. And that is through developing algorithms to distinguish between the two diseases, discovering the presence of similarities in the patterns of stages and the different types of malaria and leukemia in blood images, and reaching to solve the problem of confusion by training, verification, and testing using the mutual verification system that uses three folds. The system achieved an accuracy of 94.3% for Resnet50 and 92.3 for Alexnet in detecting and classifying the types and stages of the two diseases (malaria and leukemia). And 100% to distinguish between them.

Discussion on classification and naming scheme of fine-grained sedimentary rocks

Based on reviews and summaries of the naming schemes of fine-grained sedimentary rocks, and analysis of characteristics of fine-grained sedimentary rocks, the problems existing in the classification and naming of fine-grained sedimentary rocks are discussed. On this basis, following the principle of three-level nomenclature, a new scheme of rock classification and naming for fine-grained sedimentary rocks is determined from two perspectives: First, fine-grained sedimentary rocks are divided into 12 types in two major categories, mudstone and siltstone, according to particle size (sand, silt and mud). Second, fine-grained sedimentary rocks are divided into 18 types in four categories, carbonate rock, fine-grained felsic sedimentary rock, clay rock and mixed fine-grained sedimentary rock according to mineral composition (carbonate minerals, felsic detrital minerals and clay minerals as three end elements). Considering the importance of organic matter in unconventional oil and gas generation and evaluation, organic matter is taken as the fourth element in the scheme. Taking the organic matter contents of 0.5% and 2% as dividing points, fine grained sedimentary rocks are divided into three categories, organic-poor, organic-bearing, and organic-rich ones. The new scheme meets the requirement of unconventional oil and gas exploration and development today and solves the problem of conceptual confusion in fine-grained sedimentary rocks, providing a unified basic term system for the research of fine-grained sedimentology.

ISNet: Individual Standardization Network for Speech Emotion Recognition

Speech emotion recognition plays an essential role in human-computer interaction. However, cross-individual representation learning and individual-agnostic systems are challenging due to the distribution deviation caused by individual differences. The existing related approaches mostly use the auxiliary task of speaker recognition to eliminate individual differences. Unfortunately, although these methods can reduce interindividual voiceprint differences, it is difficult to dissociate interindividual expression differences since each individual has its unique expression habits. In this paper, we propose an individual standardization network (ISNet) for speech emotion recognition to alleviate the problem of interindividual emotion confusion caused by individual differences. Specifically, we model individual benchmarks as representations of nonemotional neutral speech, and ISNet realizes individual standardization using the automatically generated benchmark, which improves the robustness of individual-agnostic emotion representations. In response to individual differences, we also propose more comprehensive and meaningful individual-level evaluation metrics. In addition, we continue our previous work to construct a challenging large-scale speech emotion dataset (LSSED). We propose a more reasonable division method of the training set and testing set to prevent individual information leakage. Experimental results on datasets of both large and small scales have proven the effectiveness of ISNet, and the new state-of-the-art performance is achieved under the same experimental conditions on IEMOCAP and LSSED.

Postfix Form to Infix Form

Even though infix expression is easy for humans to understand, it is hard for the computers to comprehend. According to Sayan Mahapatra from GeekforGeeks , “Prefix and Postfix expressions can be evaluated faster than an infix expression. This is because we don’t need to process any brackets or follow operator precedence rule.” Using postfix expression eliminates the problem of confusion due to precedence. For example, an infix expression such as 2$3&amp;amp;4. One does not know if 2$3 should be applied first or 3&amp;amp;4 without knowing the precedence . In addition, Postfix expression can be easily calculated with the Stack base algorithm .

Class-Incremental Learning Network for Small Objects Enhancing of Semantic Segmentation in Aerial Imagery

Due to the differences in the feature distribution between classes, when the model learns in a continuous data stream, it will encounter catastrophic forgetting. The incremental learning methods have shown great potential to solve this problem. However, most existing methods based on task-incremental learning are difficult to adapt to characteristics of remote sensing scenes with few differences in appearance but large differences in features, which is not conducive to artificially distinguish task-identity document (ID). Thus, we propose a class-incremental learning (CIL) network for small objects enhancing semantic segmentation in aerial imagery. Specifically, considering the superior accuracy of the binary classifier, we propose a twin-auxiliary (TA) model that adds an auxiliary binary classification task. Then, for expansion and contraction at the edge and small object confusion problems, we introduce a diversity distillation loss, using the results of binary-classifier to constrain the multiclass segmentation results and strengthen the attention to the locations of the segmentation results that have changed. Finally, we design a conflict reduction mechanism for multihead classifier to achieve single-head prediction for CIL. Experiments demonstrate that our method has good performance on the Vaihingen and Potsdam datasets by the International Society for Photogrammetry and Remote Sensing (ISPRS), outperforming state-of-the-art (SOTA) incremental learning methods. The code will be available soon.