Short Text Research Articles

Survey Paper on Implementation of Abusive Language Detection in Chatting Application

Communication has become stronger due to exponential increase in the usage of social media in the last few years. People use them for communicating with friends, finding new friends, updating any important activities of their life, etc. Due to their growing popularity and deep reach, these mediums are infiltrated with huge Volume of spam messages. Spam message randomly sent to multiple addressees by all sorts of groups, but mostly lazy advertisers and criminals who wish to lead you to phishing sites. In this we are using various machine learning techniques for detecting spam in the short text messages and also Google vision API for detecting spam images. Key Words: Spam Filtration, Google Vision API, OWASP, Naive baise classifier, Dictionary Based Algorithm

Analysis of Deep Generative Model Impact on Feature Extraction and Dimension Reduction for Short Utterance Text-Independent Speaker Verification

Speaker verification is a biometric-based method for individual authentication. However, there are still several challenging problems in achieving high performance in short utterance text-independent conditions, maybe for weak speaker-specific features. Recently, deep learning algorithms have been used extensively in speech processing. This manuscript uses a deep belief network (DBN) as a deep generative method for feature extraction in speaker verification systems. This study aims to show the impact of using the proposed method in various challenging issues, including short utterances, text independence, language variation, and large-scale speaker verification. The proposed DBN uses MFCC as input and tries to extract more efficient features. This new representation of speaker information is evaluated in two popular speaker verification systems: GMM-UBM and i-vector-PLDA methods. The results show that, for the i-vector-PLDA system, the proposed feature decreases the EER considerably from 15.24 to 10.97%. In another experiment, DBN is used to reduce feature dimension and achieves significant results in decreasing computational time and increasing system response speed. In a case study, all the evaluations are performed for 1270 speakers of the NIST SRE2008 dataset. We show deep belief networks can be used in state-of-the-art acoustic modeling methods and more challenging datasets.

Chinese medical short text classification model based on DPECNN

Medical short text classification is of great significance to medical information extraction and medical auxiliary diagnosis. However, medical short texts face challenges such as sparse features, semantic ambiguity, and the specialized nature of the medical field, resulting in relatively low accuracy in short text classification. Taking into consideration the characteristics of medical short texts, this paper proposes a Chinese medical short text classification model based on DPECNN. First, ERNIE is utilized to learn text knowledge and information in order to enhance the model’s semantic representation capabilities. Then, the DPECNN model is employed to extract rich feature information, and the classification results are generated through a fully connected layer. In the case of DPCNN, it only considers deep-level contextual semantic information, overlooking the correlation of adjacent semantic information between channels. To address this, ECA channel attention is introduced to account for adjacent semantic information. The use of a self-normalizing activation function helps avoid the problem of vanishing gradients. To enhance the model’s robustness and generalization ability, the FGM adversarial training algorithm is employed to perturb the data. The F1 values achieved on the THUCNews, KUAKE-QIC, and CHIP-CTC datasets are 95.00%, 79.45%, and 82.81%, respectively.

Data Sorting Influence on Short Text Manual Labeling Quality for Hierarchical Classification

The precise categorization of brief texts holds significant importance in various applications within the ever-changing realm of artificial intelligence (AI) and natural language processing (NLP). Short texts are everywhere in the digital world, from social media updates to customer reviews and feedback. Nevertheless, short texts’ limited length and context pose unique challenges for accurate classification. This research article delves into the influence of data sorting methods on the quality of manual labeling in hierarchical classification, with a particular focus on short texts. The study is set against the backdrop of the increasing reliance on manual labeling in AI and NLP, highlighting its significance in the accuracy of hierarchical text classification. Methodologically, the study integrates AI, notably zero-shot learning, with human annotation processes to examine the efficacy of various data-sorting strategies. The results demonstrate how different sorting approaches impact the accuracy and consistency of manual labeling, a critical aspect of creating high-quality datasets for NLP applications. The study’s findings reveal a significant time efficiency improvement in terms of labeling, where ordered manual labeling required 760 min per 1000 samples, compared to 800 min for traditional manual labeling, illustrating the practical benefits of optimized data sorting strategies. Comparatively, ordered manual labeling achieved the highest mean accuracy rates across all hierarchical levels, with figures reaching up to 99% for segments, 95% for families, 92% for classes, and 90% for bricks, underscoring the efficiency of structured data sorting. It offers valuable insights and practical guidelines for improving labeling quality in hierarchical classification tasks, thereby advancing the precision of text analysis in AI-driven research. This abstract encapsulates the article’s background, methods, results, and conclusions, providing a comprehensive yet succinct study overview.

A systematic evaluation of text mining methods for short texts: Mapping individuals’ internal states from online posts

Short texts generated by individuals in online environments can provide social and behavioral scientists with rich insights into these individuals’ internal states. Trained manual coders can reliably interpret expressions of such internal states in text. However, manual coding imposes restrictions on the number of texts that can be analyzed, limiting our ability to extract insights from large-scale textual data. We evaluate the performance of several automatic text analysis methods in approximating trained human coders’ evaluations across four coding tasks encompassing expressions of motives, norms, emotions, and stances. Our findings suggest that commonly used dictionaries, although performing well in identifying infrequent categories, generate false positives too frequently compared to other methods. We show that large language models trained on manually coded data yield the highest performance across all case studies. However, there are also instances where simpler methods show almost equal performance. Additionally, we evaluate the effectiveness of cutting-edge generative language models like GPT-4 in coding texts for internal states with the help of short instructions (so-called zero-shot classification). While promising, these models fall short of the performance of models trained on manually analyzed data. We discuss the strengths and weaknesses of various models and explore the trade-offs between model complexity and performance in different applications. Our work informs social and behavioral scientists of the challenges associated with text mining of large textual datasets, while providing best-practice recommendations.

A semi-supervised approach of short text topic modeling using embedded fuzzy clustering for Twitter hashtag recommendation

Social media stands as a crucial information source across various real-world challenges. Platforms like Twitter, extensively used by news outlets for real-time updates, categorize news via hashtags. These hashtags act as pivotal meta-information for linking tweets to underlying themes, yet many tweets lack them, posing challenges in topic searches. Our contribution addresses this by introducing a novel heuristic for hashtag recommendation. Extracting 20 thousand tweets, 5000 each from distinct categories health, sports, politics, and technology we applied fundamental data cleaning and tokenization techniques. Leveraging Word2Vec, we vectorized tokens, capturing nuanced semantic meanings and mitigating data sparsity issues. The proposed heuristic creates clusters of different topic by combining these embedded features and idea of fuzzy C-Means technique. Develop a rule-based approach that combines both supervised and unsupervised methods to label clusters, indicating their respective topic. The experimental outcomes shows that our proposed techniques achieve better performance metrics in precision, recall, and F1-score compared to specific baseline models.

Gagarin Sixty Years Later: Earth and Place after Heidegger and Levinas

Abstract In this article I re-examine the well-known distinction between rootedness and uprootedness that Emmanuel Levinas draws in his short text “Heidegger, Gagarin and Us” (1961). This distinction addresses the relation between men and place either as an attachment to place (paganism, Heidegger) or as a freedom with regard to place (Judaism, Gagarin). I question this opposition from a contemporary perspective in environmental philosophy, namely from the growing awareness of the interconnectedness between place and Earth. I contend that this new perspective changes the understanding of dwelling today because of Earth’s exteriority with regard to place. I argue that this exteriority is neither infinite nor a totality.

Improving the term weighting log entropy of latent dirichlet allocation

<p class="AbstractText">The process of analyzing textual data involves the utilization of topic modeling techniques to uncover latent subjects within documents. The presence of numerous short texts in the Indonesian language poses additional challenges in the field of topic modeling. This study presents a substantial enhancement to the term weighting log entropy (TWLE) approach within the latent dirichlet allocation (LDA) framework, specifically tailored for topic modeling of Indonesian short texts. This work places significant emphasis on the utilization of LDA for word weighting. The research endeavor aimed to enhance the coherence and interpretability of an Indonesian topic model through the integration of local and global weights. Local Weight focuses on the distinct characteristics of each document, whereas global weight examines the broader perspective of the entire corpus of documents. The objective was to enhance the effectiveness of LDA themes by this amalgamation. The TWLE model of LDA was found to be more informative and effective than the TF-IDF LDA when compared with short Indonesian text. This work improves topic modeling in brief Indonesian compositions. Transfer learning for NLP and Indonesian language adaptation helps improve subject analysis knowledge and precision, this could boost NLP and topic modeling in Indonesian.</p>

Heterogeneous texts, hybrid genres and Anthropocene poetics: Juliana Spahr’s ‘Unnamed Dragonfly Species’ (2011) and Sarah Moss’s Summerwater (2020)

Juliana Spahr’s generically hybrid ‘Unknown Dragonfly Species’ and Sarah Moss’s composite novel Summerwater are here read in the light of their textual hybridity and with a focus on their short fiction qualities. Different in form, style and content, the two texts share a confrontation of plots concentrating on social concerns – a group of New York friends, an anonymous ‘they’, in Spahr’s ‘Unnamed Dragonfly Species’, old and young couples, families, groups of friends on holiday in Scotland in Moss’s Summerwater – with short texts (chapters in Moss, proper names of endangered species in Spahr) that point to a world beyond the human and that ironically subvert the anthropocentrism and egocentrism that drives many of the characters involved. Read against the backdrop of Ricardo Piglia’s ‘Theses on the short story’, the confrontation of different text types is shown to avoid a one-dimensional focus on human affairs and thus helps to transcend the scope each individual genre – short story, poetry or novel – might cover on its own. Looking at ‘Unnamed Dragonfly Species’ and Summerwater as textually hybrid environmental short fiction, the article argues how the genre perspective may help to produce scenarios in which the human and the more-than-human worlds permeate each other in ways that presage a poetics for the Anthropocene.

Alur Cerpen Karya Siswa Kelas XI SMA Negeri 9 Makassar

This study aims to describe the realisation of plot used in short stories by students of class XI SMA Negeri 9 Makassar. The data in this study are in the form of words, phrases, sentences in short stories, especially plot and characteristics, as the basic material for analytical studies. The source of data in this research is students' works in the form of short story texts. Data collection techniques were carried out by writing, documentation, and reading and recording. Data analysis techniques were identification, classification, analysis, and description. The results showed that the form of plot in the short stories of students of class XI SMA Negeri 9 Makassar consisted of straight plot, back-and-forth plot, mixed plot, single plot, sub-subplot plot, dense plot, loose plot, plot of fortune, plot of character, and plot of thought

From Residua to Offal: Beckett’s Counter-Anthropocene(s)

This article reflects on a series of short texts by Samuel Beckett, proposing that they centre our focus on the ‘residues’ of capitalist systems. Exemplars of the subject as victim, Beckett's human characters tend to fall prisoner to mutations in their environments; all that is left to them is the possibility of remaining, persisting amidst ruination. In this regard, Beckett’s work might be read as an act of resistance against an anthropocentric narrative of Earth’s history and against the temporalities of the Great Acceleration. This article thus identifies ‘counter-Anthropocenes’ in a selection of Beckett’s prose texts, where the transformation of the world no longer produces gain or profit, but always waste or leakage – a residue that is also ‘what remains’ or subsists, a ruin that is part of the development whereby the Capitalocene becomes a Necrocene. After an introduction to this terminology and a discussion of its resonances within a range of Beckettian ecosystems, the article then explores characters’ material modes of subsistence that resist the imperatives of the Capitalocene; they manage to ‘go on’ in a world without future. The article argues that the abundance of remains and their circulation in Beckett’s fictional worlds provide alternative narratives of human history, in which globalising and totalising perceptions can never be achieved. Rather, human beings themselves are turned into refuse caught in, and subjected to, the entropy of the Necrocene. Reading Beckett’s texts through this lens augments a conundrum at the heart of Beckett’s oeuvre and at the core of the ecological challenge of our times: the limited agency of individuals in the eco-socio-political systems that shape the everyday. Beckett’s Necrocenic poetics of loss offers us a vision of stillness and persistence against the fast-paced, production-oriented temporalities of the Anthropocene—or Capitalocene.

Employing keywords and lexical bundles within figure captions in earth science research articles

The presentation of data is considered an essential part of academic writing, especially in research articles, and it could not stand alone without a short text to describe the visual data, namely figure legends. This research aims to identify the keywords and lexical bundles frequently used in figure legends. In addition, this research also examines the functions conveyed by these linguistic features. Based on the identification of the corpus, there are 334 keywords found with the 20 highest frequencies belonging to the terminologies frequently used in earth science (fault, seismic, quartz, and formation) and words related to the visual representation of the figure legends (line/lines, map, area, figure and legend). Furthermore, noun-related bundles are found to be the greatest number for lexical bundles, with 165, followed by verb-related bundles with 79, clause-related bundles with 54, and preposition-related bundles with 17. These lexical bundles perform research-oriented, text-oriented, and participant-oriented functions. These findings suggest that lexical bundles play a role in explaining the research condition, connecting visual data and research results discussed in research articles, and displaying information about visual data or specific parts of research articles to readers.

François Gros: from antibiotics to messenger RNA.

François Gros was a prominent French Molecular Biologist who made a major contribution to the discovery of messenger RNA in 1961. He pursued outstanding research on bacterial mRNA and its translation into proteins followed by pioneering work on muscle differentiation. I was lucky to be among his graduate students and owe much of my success in science to him. In this short text I will describe how the initial post-war studies of François guided him to discover the existence of short-lived RNA in bacteria, the messenger RNA containing the information for protein synthesis. I will also recount the influence he had on his students and their carrier in science.

IMAI: A short text matching model integrating multi-angle information

The short text matching models can be roughly divided into representation-based and interaction-based approaches. However, current representation-based text matching models often lack the ability to handle sentence pairs and typically only perform feature interactions at the network’s top layer, which can lead to a loss of semantic focus. The interactive text matching model has significant shortcomings in extracting differential information between sentences and may ignore global information. To address these issues, this article proposes a model structure that combines a dual-tower architecture with an interactive component, which compensates for their respective weaknesses in extracting sentence semantic information. Simultaneously, a method for integrating semantic information is proposed, enabling the model to capture both the interactive information between sentence pairs and the differential information between sentences, thereby addressing the issues with the aforementioned approaches. In the process of network training, a combination of cross-entropy and cosine similarity is used to calculate the model loss. The model is optimized to a stable state. Experiments on the commonly used datasets of QQP and MRPC validate the effectiveness of the proposed model, and its performance is stably improved.

Taming Misinformation: Fake Review Detection on Social Media platform using Hybrid Ensemble Technique

In today's digital world, we witness exponential growth in the generation of textual content on a daily basis. However, the widespread dissemination of information through social media, online forums, and news websites has given rise to the proliferation of fake views, opinions, and reviews, posing a significant challenge in the battle against misinformation and manipulation. Machine Learning has become increasingly integral to real-world online activities, particularly in the area of Artificial Intelligence. Traditional methods often struggle to keep pace with the relentless creation of internet data. Consequently, short text processing has emerged as a new domain for the application of Machine Learning. This is where sentiment analysis come to the forefront, offering potent tools for discerning the authenticity of online content. Detecting and combating these fabricated sentiments are crucial for preserving the integrity of information and ensuring informed decision-making. This work focus on the previously unexplored area of user comments on review data. By leveraging N-gram technique and hybrid ensemble classification approaches, the research addresses critical issues in fake reviews classification, and sentiment analysis. The aim of this work is to detect fake reviews with limited text using NLP feature extraction, and hybrid ensemble classification algorithms, ultimately contributing to the enhancement of information integrity and decision-making in the digital age.

Cognitive impairment assessment through handwriting (COGITAT) score: a novel tool that predicts cognitive state from handwriting for forensic and clinical applications.

Handwriting deteriorates proportionally to the writer's cognitive state. Such knowledge is of special importance in the case of a contested will, where dementia of the testator is claimed, but medical records are often insufficient to decide what the testator's cognitive state really was. By contrast, if the will is handwritten, handwriting analysis allows us to gauge the testator's cognitive state at the precise moment when he/she was writing the will. However, quantitative methods are needed to precisely evaluate whether the writer's cognitive state was normal or not. We aim to provide a test that quantifies handwriting deterioration to gauge a writer's cognitive state. We consecutively enrolled patients who came for the evaluation of cognitive impairment at the Outpatient Clinic for Cognitive Impairment of the Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics and Maternal and Child Sciences (DINOGMI) of the University of Genoa, Italy. Additionally, we enrolled their caregivers. We asked them to write a short text by hand, and we administered the Mini Mental State Examination (MMSE). Then, we investigated which handwriting parameters correlated with cognitive state as gauged by the MMSE. Our study found that a single score, which we called the COGnitive Impairment Through hAndwriTing (COGITAT) score, reliably allows us to predict the writer's cognitive state. The COGITAT score may be a valuable tool to gage the cognitive state of the author of a manuscript. This score may be especially useful in contested handwritten wills, where clinical examination of the writer is precluded.

Contextualized Satire Detection in Short Texts Using Deep Learning Techniques

Satire is prominent in user-generated content on various online platforms in the form of satirical news, customer reviews, blogs, articles, and short messages that are typically of an informal nature. As satire is also used to disseminate false information on the Internet, its computational detection has become a well-known issue. Existing work focuses primarily on formal document- or sentence-level textual data, whereas informal short texts have gotten less attention for satire detection. This paper presents a new model called BiLSTM self-attention (BiSAT) for detecting satire in informal short texts. It consists of various components such as input, embedding, self-attention, and two bi-directional long short-term memory (BiLSTM) layers for learning crucial contextual information pertaining to the satire present in the texts. The input layer uses the text as input to create an input vector, which is then given to the embedding layer to create the appropriate numeric vector. The output of the embedding layer is passed on to the first BiLSTM layer, which extracts contextual information-based sequences in the opposite direction. Between the first and second BiLSTM layers, a self-attention layer is employed to draw attention to the important satirical information that is acquired by the hidden layer of the first BiLSTM. The BiSAT model also takes a classic feature engineering approach, employing a 13-dimensional auxiliary feature vector comprised of features from four separate feature categories: sentiment, punctuation, hyperbole, and affective. The proposed BiSAT model is empirically evaluated on two benchmark datasets and a newly created dataset called Satire-280. It outperforms existing research and baseline methods by a significant margin. The Satire-280 dataset along with code can be downloaded from GitHub repository: https://github.com/Ashraf-Kamal/Satire-Detection.

Development of GOFIN (Google form interactive)-based on the short story worksheet with differentiated learning approach

The availability of interactive-based student worksheets that also take into account differences in learning styles is necessary to address the challenges of the digital era in teaching short story writing. This research aims to develop Student Worksheets (LKPD) based on Gofin (Interactive Google Form) for short story text learning with the application of a differentiated approach that is suitable for interests, abilities, and learning styles. This developmental research adapts the ADDIE model and carries out three steps, namely analysis, design, and development. The implementation and evaluation steps are conducted in subsequent research. The research results indicate that this LKPD is highly suitable and engaging with a material validity rate of 87%, media validity of 83%, and student response of 96%. This LKPD is designed to accommodate differences in students' learning styles (kinesthetic, auditory, visual) so that learning objectives can be maximally achieved.

A randomised controlled trial to evaluate the effectiveness of a culture and disease-specific, patient-centric multi-component tobacco cessation intervention package for the patients attending non-communicable disease clinics in Punjab, India

Background Developing an infrastructure to support tobacco cessation through existing systems and resources is crucial for ensuring the greatest possible access to cessation services. The present study aims to evaluate the effectiveness of a newly developed multi-component cessation among tobacco users in Non- Communicable Disease (NCD) clinics, functioning under the National Programme for Prevention & Control of Cancer, Diabetes, Cardiovascular Diseases, & Stroke (NPCDCS) of the Government of India. Methods The intervention package consisting of culture- and disease-specific four face-to-face counselling sessions, pamphlets, and short text messages (bilingual) with follow-ups at 3rd, 6th, and 9th months with an endline assessment at 12th months was delivered to the intervention arm of the two-arm- parallel group randomised controlled trial at two selected NCD clinics. Self-reported seven-day abstinence, frequency of use, expenditure in seven days at each follow-up, FTND score, stage of change and plasma cotinine values were assessed at baseline, follow-ups, and endline (using Liquid Chromatography –Mass Spectrometry), as applicable. Results The intervention arm reported a significantly more reduction in self-reported frequency of tobacco use at 6 months (mean: 13.6, 95% CI (7.8–19.4)), 9 months (mean: 20.3, 95% CI (12.2–28.4)) and 12 months (mean: 18.7, 95% CI (8.7–28.7)). The plasma cotinine concentration at endline in the intervention arm was statistically significantly lower than the baseline concentration. Conclusion Strengthening existing health systems is crucial for offering cessation support in the resource-restraint setting of LMICs to assist in quitting sustainably.

Supplementing a Paper Questionnaire with Web and Two-way Short Message Service (SMS) Surveys

Abstract When deciding which modes to offer, researchers consider cost, known respondent contact information, and potential mode effects. For a short survey on employment, we evaluate the effect of adding one of two new electronic data collection modes to a mailed questionnaire. We sent a survey to principals who previously responded to the National Center for Education Statistics’ (NCES) National Teacher and Principal Survey (NTPS) asking about their current job status. This questionnaire, known as the Principal Follow-up Survey (PFS), has typically been administered as a short paper form that is mailed to NTPS respondents. In 2022, the PFS introduced two new modes of completion, and principals were randomly assigned to receive: (i) a paper form only; (ii) a paper form, as well as emails with a direct link to complete a web survey; or (iii) a paper form, as well as invitations by text message to complete an automated two-way short message service text survey by responding to texted “yes/no” questions. This article compares overall response rates and time-to-response by mode to determine respondent preferences for completing short surveys. Adding either electronic mode significantly increased response rates and decreased the number of days in which completed surveys were received, compared to offering only a paper questionnaire. Although email and text messages are both forms of electronic communication that may be accessible on a smartphone, the added text message survey resulted in higher response rates than the added web survey. This suggests that respondents interact differently with emails and text messages they receive and that offering an option to complete a survey by text message can increase the speed and efficiency of data collection for short surveys.