Aspects Of Classification Research Articles

Aspect Category Detection Using Bi-LSTM-Based Deep Learning for Sentiment Analysis for Hindi

Aspect Category Detection Using Bi-LSTM-Based Deep Learning for Sentiment Analysis for Hindi

DEEP LEARNING APPROACH FOR ASPECT CATEGORY DETECTION: A BIBLIOMETRIC ANALYSIS

This article presents a quantitative and qualitative assessment of current research trends by conducting a bibliometric analysis of the sentiment analysis literature from 2020 to March 2024 using the Scopus database. Our focus is on the review of scientific documents, the arrangement of subject categories, the research trend in aspect category detection, the top 10 scholars that write the most number articles in aspect category detection and keyword trends. Our research shows that specialists in computer science, engineering, mathematics, medicine, decision science, material science, social sciences, business and management accounting, energy, and health are the most common topic groups in this industry. A study of keywords shows that terms like “BERT” and “deep learning” are frequently used together. This highlights the use of sophisticated models like BERT in this field and suggests a tendency to use innovative architecture to achieve better results. Conversely, although terms such as “sentiment analysis” and “aspect-based sentiment analysis” have modest frequency, their link strengths indicate a significant correlation with the main theme, emphasising the relationship between aspect category detection and sentiment analysis in research projects. we also provide the deep-learning technique used by the author for aspect category detection.

"Urgency" and "emergency" in the system of law-making proccess: conceptual and categorical dimension

It is indicated that the course of events at the global and regional levels of social development during recent years is difficult to characterize unambiguously from the standpoint of purely positive or negative aspects. However, it can be unequivocally noted that modern socio-political processes at the global and regional levels have a particularly sensitive effect on legal phenomena and processes. In the article, the author focuses on the relevance of the study of the conceptual and categorical foundations of the categories «urgency» and «emergency» and the peculiarities of their manifestation within the framework of law­making process. According to the author of the article, the achievement of the above is due to the need to carry out a comprehensive and conceptual analysis through the prism of scientific views and approaches of scientists. Attention was drawn to the fact that in connection with numerous factors of a subjective and objective nature, in particular, the duration of the war on the territory of Ukraine, the spread of the SARS- CoV-2 virus, the development of technologies, especially in the information field, the issues of new mechanisms for the development and adoption of the system are being updated normative legal acts taking into account the existing requirements and realities of the modern state of development of society, law and the state. Analysis of doctrinal approaches to this issue will provide clarification of the conceptual and categorical aspects of the relevant legal categories. It is substantiated that in cognitive terms it is expedient to analyze these legal categories through the prism of scientific views and concepts, which will allow generalizing their properties and improving their definitions. The content of the work includes a doctrinal analysis of the concepts of «urgency» and «emergency» as special legal categories. It was concluded that the categories «emergency» and «urgency» within the framework of law-making have a high level of relevance; require thorough, comprehensive scientific research within the framework of legal theory; are characterized by the presence of a common purpose, essential differences.

Enhancing aspect category detection in imbalanced online reviews: An integrated approach using Select-SMOTE and LightGBM

Aspect category detection (ACD) is a pivotal subtask within the field of sentiment analysis in natural language processing, aiming to identify implicit aspect category information in online review texts. In real-world scenarios of online review category detection tasks, data imbalance often arises, leading to skewed distributions among distinct review categories. This phenomenon poses substantial challenges for accurately recognizing minority categories through modeling. To address this, we propose a method for detecting imbalanced aspect categories by combining the selective synthetic over-sampling (Select-SMOTE) algorithm with the light gradient boosting machine (LightGBM). Our approach commences with text data representation through features, followed by a strategy involving joint sample partitioning and boundary optimization within the feature space to generate minority class samples. This partitioning strategy aligns generated data more closely with the original distribution, while the boundary optimization module enhances classification performance by eliminating samples near boundaries. Subsequently, the balanced dataset is input to the LGB model, enabling the extraction of aspect category information through parameter optimization and class weight assignment. Finally, our method is evaluated using the SemEval and SentiHood datasets and compared with prevailing sampling methods and classification models. Empirical results manifestly demonstrate the method’s superiority across diverse metrics, reflecting robustness and effective mitigation of imbalanced data challenges in ACD.

Dual edge-embedding graph convolutional network for unified aspect-based sentiment analysis

Aspect-based sentiment analysis (ABSA) mainly involves three elements, i.e., aspect, target, and sentiment. Most existing research focuses on aspect-sentiment or target-sentiment separately, while connection among the three elements is still underexploited. In order to extract three interrelated elements from a sentence simultaneously, we propose a unified end-to-end model, which is based on one aspect via a virtual edge attention mechanism and a graph convolutional network (GCN) to detect related targets and to analyze the corresponding sentiment polarity. The virtual edges are decided by the association intensity between the aspect and the words of the sentence. Instead of naively regarding all connecting edges as binary values, we design a novel GCN with a dual edge-embedding method to help information flow among each word in the sentence and between the aspect category and the sentence. Extensive experiments on two open datasets show that our model outperforms the state-of-the-art models in the aspect-target-sentiment triples detection task and its two subtasks.

Exploring large language models for the generation of synthetic training samples for aspect-based sentiment analysis in low resource settings

Aspect-Based Sentiment Analysis (ABSA) is a fine-grained task in sentiment analysis, aiming to identify sentiment expressed towards specific aspects of an entity. This paper explores the use of Large Language Models (LLMs), specifically GPT-3.5-turbo and Llama-3-70B, for generating annotated data in Aspect-Based Sentiment Analysis (ABSA), aiming to address the scarcity of labeled datasets in the field. Two low-resource scenarios are considered, with 25 and 500 manually annotated examples available. In the 25-example scenario, adding synthetic examples generated through few-shot prompting resulted in F1 scores of 81.33 for Aspect Category Detection (ACD) and 71.71 for Aspect Category Sentiment Analysis (ACSA). For the 500-example scenario, synthetic data augmentation showed a notable gain only for the ACSA task, raising the F1 score from 84.54 to 86.70.

The Path Towards Personal Growth in Spousal Caregiving

Caregiver support is necessary for many long-term ill or disabled people to enable them to live at home. However, the impact of this care relationship on the caregiver is still partly unknown. This article focuses on the process of achieving personal growth in spousal caregiving. The aim of our study was to conduct an integrative literature review to describe the aspects associated with achieving personal growth as well as to analyze the relationship between the aspects involved. The thematic analysis method was used with inductive content classification to modify the themes. The analysis highlighted three main aspect categories connected with achieving personal growth: the means, coping styles, and personal or caregiving resources. In addition, the path from adversities and challenges towards personal growth and positive experiences in spousal caregiving was revealed by analyzing the relationship between the identified aspects. We found that the means, based on social support and self-care, help caregivers to adopt positive coping strategies, reframing, and resilience, which are associated with achieving personal growth. Also, personal or caregiving resources, such as meaningful caring experiences, religiosity, and spirituality, hope, positive personality characteristics, family functioning, and disease severity of the care recipient, were associated with achieving growth. Finally, our findings suggest that it is important to further study the path towards personal growth in spousal caregiving and to find the best possible ways to support caregivers in difficult moments and periods of caregiving.

Aspect-Level Sentiment Analysis on GoPay App Reviews Using Multilayer Perceptron and Word Embeddings

The increase of smartphone usage in Indonesia has encouraged the development of digital wallet applications, one of which is GoPay. Nowadays, GoPay has gained significant popularity among the public in Indonesia. Therefore, this research conducts aspect-level sentiment analysis to analyze user reviews of the GoPay application in more detail and depth. The sentiment analysis process in this study utilizes the Multilayer Perceptron (MLP) with fastText and word2vec as word embeddings. The dataset used is GoPay application reviews, which consist of 15,000 reviews collected from Google Play Store. The dataset is categorized into three main aspects: Feature and functionality, App Interface, and User Satisfaction. The stages of the research include data preparation, data preprocessing, word embeddings, model training, and model testing and evaluation. This research explores the effect of fastText and word2vec as word embeddings on model performance. Furthermore, this research examines the application of oversampling techniques, such as SMOTE and Random Oversampling. Based on the experiments conducted, utilizing fastText as word embeddings in MLP with a balanced dataset resulted the best model performance, with an F1-Score of 97%, Recall of 96%, and Precision of 95% for aspect category classification. Then, for sentiment classification, using fastText on MLP with a balanced dataset resulted in a value of 98% for each of the F1-score, Recall, and Precision metrics. This research validates that MLP is effective for aspect-level sentiment analysis, delivering strong evaluation results.

Versorgungsqualität und -sicherheit nach Implantation eines Linksherzunterstützungssystems: eine qualitative Studie zur Patient*innenperspektive auf relevante Versorgungsaspekte

IntroductionImplantation of a left ventricular assist device (LVAD) requires extensive aftercare. It is largely unclear how aftercare should be designed from the patients’ perspective. Implications can be developed based on an examination of the healthcare context. Its main components are mapped on five tiers in the Human Factors of Home Health Care Model by Henriksen, Joseph, and Zayas-Caban (2009). Using this model, the present study explores the patient perspective on the context of healthcare after an LVAD implantation. MethodsWe employed a qualitative cross-sectional study, in which LVAD patients participated in semi-structured interviews. The transcribed interviews were analyzed using content analysis. First, relevant meaning units were identified and deductively categorized into the model. Then, categories of care-related aspects were developed inductively within each of the model tiers. ResultsWe interviewed 18 patients aged 33 to 78 years who had been living with the LVAD between a few weeks and more than 10 years. Twenty-eight categories related to care aspects were developed within the model tiers: 3 categories on patient characteristics (e.g., self-management skills), 3 on caregiver characteristics (e.g., professionalism), 11 healthcare-related tasks and requirements (e.g., wound management), 8 on factors of the physical environment (e.g., controllability), medical devices and technologies (e.g., carrying systems for external components), and cultural, social and community environment (e.g., interaction with peers), as well as 3 on external environmental factors (e.g., healthcare infrastructure). DiscussionThe present study represents the first investigation focusing on aspects of the healthcare context influencing healthcare quality and safety from the perspective of LVAD patients in Germany. LVAD aftercare covers a broad and complex range of tasks. For this, patients, caregivers and healthcare professionals need specific knowledge, which is lacking in various respects. In the first place, this is compensated by the patients’ own initiative and the personal care provided by the VAD outpatient clinics. ConclusionThree key recommendations to optimize aftercare from the patient perspective are derived: Patients would benefit from a more flexible and decentralized aftercare concept, to which telemedicine could contribute. LVAD-specific expertise among general healthcare providers is perceived as insufficient by patients and could be strengthened through training and counseling services. The broad scope of tasks and the high level of responsibilities in LVAD aftercare pose challenges for patients and their families, which could be addressed through continuous information and training programs.

AUTISM IN ADULTS: Feelings and changes after a late diagnosis

Previously considered a rare childhood disorder, autism today is a frequent diagnosis with increasing demand in all age groups, including people who were not diagnosed in childhood, but who perceive characteristics of this condition. This study, which is part of a broader research on the experience of Brazilian adults who were diagnosed with autism spectrum disorder (ASD) after 18 years, aims to explore the feelings and life changes of this group of people after the diagnosis. To this end, a qualitative research was carried out in which 12 participants were interviewed and the data were analyzed using the content analysis method, in its categorical aspect. Considering the objective of this work, in the analysis of the interviews, these categories emerged: alexithymia, relif, revisiting the past, self-compassion and acquired rights. As a result, relief was observed as a feeling of great relevance and predominance, accompanied by a process of understanding the past and self-compassion and self-respect. It was concluded that the late receipt of the diagnosis of autism by this group of participants allowed a better understanding of their characteristics and behaviors, reducing the pressure they exerted on themselves, in addition to enabling access to rights guaranteed by law.

EAFL: Equilibrium Augmentation Mechanism to Enhance Federated Learning for Aspect Category Sentiment Analysis

Aspect Category Sentiment Analysis (ACSA) involves identifying sentiment categories for specific aspects of a sentence. Despite the progress made in pre-trained language models for extracting semantic and categorical representations, the sharing of sensitive data remains challenging due to data privacy protection. Federated Learning (FL) offers a solution by enabling decentralized model training without accessing raw datasets. However, isolated data encounters difficulties with text heterogeneity, which is crucial for aspect-based sentiment analysis (ABSA) subtask like ACSA. To address these problems, this paper presents a novel method called the Equilibrium Augmentation Mechanism to Enhance Federated Learning for Aspect Category Sentiment Analysis (EAFL-ACSA). The EAFL mechanism seamlessly integrates FL’s data privacy preservation capabilities with a variety of augmented data representations. Our unified model safeguards data privacy through FL and incorporates multiple features through data augmentation to enhance encrypted data. Furthermore, the parameter-efficient tuning mechanism incorporated in our framework enhances model performance and generates fine-tuned augmented sentences. Extensive experiments on five benchmark datasets demonstrates that EAFL consistently outperforms the best models, achieving a competitive level of performance while protecting data privacy. The equilibrium data augmentation mechanism not only advances aspect category detection but also enhances semantic data representations, effectively addressing critical concerns surrounding data privacy preservation within a unified framework.

Using bioacoustics tools to clarify species relationships: vocalization of Cycloramphus lithomimeticus (Anura Cycloramphidae), with a phylogenetic comparative analysis of genus based on call

In anurans communities, vocalizations play a fundamental role in social interactions, being the most important behavioral modality of communication. Therefore, the description of acoustic signals is indispensable for understanding aspects of their evolutionary history. The Cycloramphus genus consists of 30 species allocated into five groups, of which only 19 have the vocalization described. Cycloramphus lithomimeticus is a small-sized frog belonging to the C. granulosus group. This poorly known species is associated with rock surfaces near waterfalls in Atlantic montane forests, restricted to the state of Rio de Janeiro, southeastern Brazil. Currently, it’s classified as Near Threatened species category of IUCN and many aspects of its natural history remain unknown. Herein, we describe the advertisement call of C. lithomimeticus extending its known geographical distribution and providing additional data on the natural history of the species. We recorded and analyzed 44 calls from 10 males of C. lithomimeticus. The advertisement call of C. lithomimeticus consists of a single pulsed note with 0.3–0.8 sec of duration, emitted sporadically, with an interval between calls of 27.5–226.3 sec, composed by 7–17 pulses emitted at a rate of 18–26/sec, with a peak frequency of 2062.5–3000 Hz. After defining the correspondence between structures for systematic purposes, we made a comparative analysis determined by establishing homology connections with other species of the genus. According to our dominant frequency evolutionary reconstruction, an evolutionary increase in frequency mainly occurred in the C. lithomimeticus species lineage from an ancestral dominant frequency of 1500 Hz of the clade with C. eleutherodactylus species group, C. granulosus species group, and C. fuliginosus species group. This study contributes to our knowledge about the biology of C. lithomimeticus, and our results provide basic data for further acoustic, taxonomic and ecological studies in the genus Cycloramphus.

Aspect category sentiment analysis based on multiple attention mechanisms and pre-trained models

The rapid rise of e-commerce platforms has changed people's shopping habits, driving the popularity of online shopping. Users express their opinions on products and services by purchasing products on platforms and posting comments. These comment data contain rich user experiences, which are crucial for enterprises to understand user needs and improve product quality. Sentiment analysis of comment text is an important research direction in text mining, focusing on how to extract user evaluations of products from comment data to provide comprehensive, authentic, and accurate product feedback. This paper mainly investigates aspect-based fine-grained sentiment analysis. Because users express multiple sentiments towards different aspects in comments, coarse-grained sentiment analysis cannot accurately capture users' sentiment tendencies. This study utilizes artificial intelligence technology based on deep learning, firstly, it constructs auxiliary training samples to transform aspect sentiment tasks into machine reading comprehension or language inference tasks, using BERT model to extract text features and sentence features from comment data. To guide the model to focus on the features most relevant to the given aspect, a cross-attention mechanism is utilized to cross-focus the features of text with aspect category features. Finally, the sentiment polarity of given aspect category text can be predicted through a forward network. Experimental results on multiple datasets demonstrate that this method outperforms other deep learning models.

Aspect category sentiment analysis based on multiple attention mechanisms and pre-trained models

The rapid rise of e-commerce platforms has changed people's shopping habits, driving the popularity of online shopping. Users express their opinions on products and services by purchasing products on platforms and posting comments. These comment data contain rich user experiences, which are crucial for enterprises to understand user needs and improve product quality. Sentiment analysis of comment text is an important research direction in text mining, focusing on how to extract user evaluations of products from comment data to provide comprehensive, authentic, and accurate product feedback. This paper mainly investigates aspect-based fine-grained sentiment analysis. Because users express multiple sentiments towards different aspects in comments, coarse-grained sentiment analysis cannot accurately capture users' sentiment tendencies. This study utilizes artificial intelligence technology based on deep learning, firstly, it constructs auxiliary training samples to transform aspect sentiment tasks into machine reading comprehension or language inference tasks, using BERT model to extract text features and sentence features from comment data. To guide the model to focus on the features most relevant to the given aspect, a cross-attention mechanism is utilized to cross-focus the features of text with aspect category features. Finally, the sentiment polarity of given aspect category text can be predicted through a forward network. Experimental results on multiple datasets demonstrate that this method outperforms other deep learning models.

The identities of employed students: Striving to reduce distinctiveness from the typical student

AbstractEndorsement of the employed student identity can provide social support for employed students or protection from negative intergroup comparisons. However, not much is known about what identity aspects or characteristics comprise the employed student identity and how they become important and central to that identity. Using data from 215 employed university students in the UK, we investigated two research questions (RQ's) in this mixed‐method study. RQ1. What are the identity aspects that participants ascribe to the employed student identity? RQ2. Are identity aspects that distinguish employed from non‐employed students, and are considered more suitable for employed versus non‐employed students, more central and more important to the employed students’ self‐concept? A thematic analysis categorized the identity aspects that participants self‐generated into 14 distinct categories, with the most important categories being hard‐working, being organized, having motivation, and discipline. Multilevel analyses of identity aspects within individuals revealed that distinctiveness was negatively associated with the importance and centrality of aspects, whereas suitability for employed students was positively associated with the importance and centrality of aspects. We offer practical value through revealing important identity aspects which inoculate employed students against negative intergroup comparisons, and theoretical value through suggesting future avenues for employed students’ identity construction.Public Significance StatementBeing an employed student carries a social cost but defining as one can help in obtaining social support or protecting from negative comparisons with non‐employed students. We identify fourteen categories of aspects that can define oneself as an employed student (e.g., motivation, hard work) with employed students experiencing those aspects which are more suitable to them as more identity‐defining, and those aspects which differentiate them from non‐employed students as less identity‐defining.

Non-Verbal Communication Aspect in Paper Umbrella Craft as Intangible Cultural Heritage in Malang

Paper umbrellas are one of the crafts included in the intangible cultural heritage of Malang, East Java. Currently, it is known that there is only one group of craftsmen who still produce this paper umbrella craft. Although not many of them, this paper umbrella craft still exists among people who like traditional handicrafts. This is evident from how often paper umbrella crafts appear at craft exhibitions in cities in Indonesia. Consequently, there is a pressing need for in-depth research on paper umbrella craftsmanship in Malang City. A valuable avenue for exploration is the examination of non-verbal communication aspects inherent in paper umbrella craftsmanship. This study aimed to determine the non-verbal communication aspects contained in paper umbrella crafts. The research employs a descriptive qualitative approach, focusing on analyzing non-verbal communication aspects. Data was collected using unstructured interview methods, documentation, and field surveys. The collected data was reduced for analysis using nonverbal communication theory. The study findings are presented descriptively, detailing the observed non-verbal communication aspects. The findings reveal three distinct categories of non-verbal communication aspects in paper umbrella craftsmanship: fixed features, semi-fixed features, and non-fixed features. Fixed features include shapes depicted on the umbrella. Semi fixed feature is a choice of material used to make paper umbrellas, colors, and motifs. The materials used are bamboo and paper. The colors of the umbrellas are red and green. The motif drawn on the umbrella is a floral motif. Flowers are present because they are the favorite shapes of the craftsmen who make umbrellas. Non-fixed features encompass the entire process of crafting paper umbrellas, from material preparation to final finishing touches. Keywords: umbrella paper, craft, tangible cultural heritage

Система обучения категории вида русских глаголов в китайской аудитории

The aim of the study is to propose a system of teaching the category of aspect of Russian verbs to Chinese students. The paper presents a research conducted at the Altai State Pedagogical University, which helped to identify the main problems in the assimilation of types of verbs by Chinese students and to test the developed teaching system, which made it possible to reduce the number of errors in the use of the grammatical category of aspect by three times for each specific Chinese student who took part in the research. The scientific novelty of the study lies in developing the system of lessons to overcome linguistic difficulties in the use of the category of aspect by Chinese students in the framework of teaching Russian as a foreign language. As a result, the teaching system is presented that may be useful for teachers of Russian as a foreign language, methodologists, and linguists interested in studying language transfer, aspect in Russian and Chinese.

The Compromised Tense-Aspect Distinction in Bantu Languages: The Case of Nyakyusa

This article examined the compromised tense-aspect distinction in Bantu languages, using Nyakyusa as a case study. The review of scholarly works indicates that, in Bantu languages, the changing nature of tense-aspect systems remains to be the source of the problem for tense-aspect distinction. The paper investigates the source of the problem and a way in which Nyakyusa can describe tense-aspect distinction. The study was conducted in Kyela District in Mbeya Region where many native speakers live. Data were collected through narrative stories, interviews and written texts. The findings reveal that although the -ile suffix is the source of the compromised tense- aspect distinction as it assumes different roles; it remains a good candidate for testing tense-aspect distinction. When the - ile suffix co- occurs with the pre-root formative, -a- it marks the past tense whereas the suffix -ile occurring alone marks aspect categories. Also, when the negative marker -ka- is introduced the -ile suffix disappears for aspect and remains for tense.

On the development of the substantive form of the Russian verb (the problem of the verb aspect in -(e)nie, -tie word forms in the 18th–early 21st c.)

The substantive form of the verb in the Russian language includes -nie, -enie, and -tie word forms. There is a historical tendency towards the grammaticalization of these formations. The category of aspect penetrates into their composition through the -yvanie forms, preserving the meaning of the imperfective aspect. Their productivity has been growing since the middle of the 18th c. by “waves,” reaching a maximum in 1761–1800, 1861–1880, 1921–1940, and 2001–2020. Half of the -yvanie forms are formed earlier than their pairs from the perfective verbs ending in -(e)nie and -tie which combine the meaning of both types; in the synchronously operating system, the aspectual pair is “completed.” With pure aspectual prefixes, po- and s-, the -yvanie forms are practically not formed. If the aspectual opposition does not add up at the stage of perfectivization, perfective prefix verbs easily form at the stage of imperfectivization their aspectual pair ending in -yvanie. Thus, the category of aspect penetrates into the system of the substantive form of the verb, enhancing the effect of this verbal category on the nominal forms of the verb.

Aspect-Level Sentiment Analysis of Customer Reviews Based on Neural Multi-task Learning

In the era of big data, major e-commerce platforms are facing the challenge of an exponential growth in the number of user comments. Effectively utilizing these comments has become an urgent issue. Traditional manual statistical methods are no longer able to meet the demands for accuracy and real-time analysis. The rise of artificial intelligence-based text mining techniques provides a new approach to address this problem. By building deep learning analysis models, it is possible to uncover user preferences and product characteristics, helping businesses proactively adjust sales strategies, improve product and service quality, and achieve precise marketing. Text sentiment analysis is one of the important research directions in the field of text mining. Its basic idea is to transform subjective texts with emotional color into structured data, and then use machine learning, deep learning, and other artificial intelligence technologies to extract emotional features and discover knowledge patterns. This article proposes a deep learning-based sentiment analysis model, which includes a shared sentiment prediction layer used to transfer emotional knowledge between different aspect categories and alleviate the problem of insufficient data. The model consists of two parts: the Aspect Category Detection (ACD) classifier based on attention mechanism and the Aspect Category Sentiment Analysis (ACSA) classifier. The ACD part generates word weights using attention mechanism, while the ACSA part predicts the sentiment of words and combines weights to determine the sentiment of aspect categories within sentences.