Neural Text Normalization for Turkish Social Media

In this digital era, college students have become familiar with social media such as Facebook, twitter, and Instagram. For English Letters Students, they also use English when updating their status, tweeting their thought, or giving captions in their photos. That phenomenon results in the students’ interest in analyzing language use in social media, proven by some theses examining social media as the data. It means that the students have started to observe English language and literacy as social practice, for example is when a student observed English Letters students’ grammatical errors in their social media accounts. That fact reveals the student’s awareness of the importance of English language and literacy as social practice. A questionnaire was distributed to 64 students who take Language Research Methodology Class. Most of them access Instagram and Facebook in their daily life and spend half of their day accessing the media. Their activities involve reading posts on information and observe comments in the form of sentences. Pedagogical implication of the students’ responses was teachers should bring social issues which happen in society in which the students live to build their critical thinking. Therefore, the students are involved in social practice while developing their critical literacy.

This article discusses the function of language and pragmatic meaning in social media to combat Covid-19. Theoretical references used are ethnopragmatics. Ethnographic studies basically examine language from a cultural aspect. Cultural aspects include knowledge, behavior, and artifacts. Meanwhile, pragmatics is a study of language use on the basis of context. This research uses data sources from Instagram, Facebook, WA, and email. Research data in the form of language use in social media above. Data from Instagram and Facebook are downloaded from the internet, while data from email is collected through the direction of researchers with guidance on the instrument. Data collection techniques are done through reading and taking notes. Concrete data analysis techniques are carried out through data identification, data classification, data interpretation, and reporting in the form of journal articles. The findings of research on language subfunctions include asking, inviting or influencing, being sarcastic, motivating and criticizing, and saying thank you. Meanwhile, the pragmatic meaning used in social media includes indirectly advising, persuading, indirectly reminding, subtly insinuating, inviting, complaining, and give advice.

This study provides a brief state-of-the-art overview of sociolinguistic research in Saudi Arabia. It begins with a historical review of the language situation in Saudi Arabia in an attempt to relate the past situation with current language use, focusing on studies that examine Arabic dialects of major tribes and urban centers in the Arabian Peninsula before and after the creation of the Kingdom of Saudi Arabia. It continues with a schematic survey of major trends in Saudi sociolinguistic works over the past few decades, including language variation and change, language contact, language attitudes, and code-switching. As such trends unfolded, researchers investigated the processes of dialect contact, mixing, and, ultimately, koineization and followed, in particular, the socioeconomic changes that Saudi Arabia has witnessed in terms of the spread of education, population movements, and urbanization patterns. Special attention is given to the various sociolinguistic methods employed in such works, as well as the linguistic features that are examined. This study concluded with future directions for Saudi sociolinguistic research in an effort to highlight topics and concerns that need further research, such as certain sociolects and ethnolects, and also new topics and trends associated with language use in social media, particularly by the younger generation.

#Sleepyteens: Social media use in adolescence is associated with poor sleep quality, anxiety, depression and low self-esteem

The symbolic collection technology has greatly simplified information preservation in social media, resulting in an overwhelming amount of users’ “favorites” collections. Digital hoarding, a new problematic use in hedonic social media, has gradually appeared as a serious phenomenon nowadays. This study aims to investigate the mechanism underlying digital hoarding in the hedonic context, exploring the causes of digital hoarding and how it is exacerbated in social media. First, we draw on the conceptual lens of Fear of Missing Out (FoMO) to understand the generation mechanism of digital hoarding in social media. Meanwhile, we consider the moderating effect of social media affordances between the relationship of FoMO and digital hoarding from a human-information interaction perspective. Our preliminary study collected 233 valid questionnaires and tested the hypotheses by partial least squares structural equation modeling. The tentative findings show that FoMO has a significant positive effect on digital hoarding, which is strengthened by content sharing affordance. Moreover, information anxiety and attachment anxiety can strongly predict FoMO. Finally, we discuss the potential contributions and future directions of this work.

Despite being a fairly recent phenomenon, emojis have quickly become ubiquitous.Besides their extensive use in social media, they are now also invoked in customer surveys and feedback forms.Hence, there is a need for techniques to understand their sentiment and emotion.In this work, we provide a method to quantify the emotional association of basic emotions such as anger, fear, joy, and sadness for a set of emojis.We collect and process a unique corpus of 20 million emoji-centric tweets, such that we can capture rich emoji semantics using a comparably small dataset.We evaluate the induced emotion profiles of emojis with regard to their ability to predict word affect intensities as well as sentiment scores.

In the contemporary era, artificial intelligence is becoming increasingly advanced, especially with its use in social media through fake bot accounts as a means for propaganda. This research looks into the Russian AI-operated bot farms and how they exacerbated the political disinformation campaigns during the U.S. election debate in 2024. The focus is on how these bots are said to generate and escalate disinformation through the computational propaganda theory. Using the case study method, the authors analyze an array of secondary sources, expert interviews, and statistical data. The results of the study reveal that bot farms impede political debates through active dissemination of misinformation. In addition, the study presents more effective ways to enhance media and disinformation detection literacy.

The rise of digital communication via hand phone and Internet has led to the widespread use of short-form words and abbreviations in text messaging. This trend poses challenges for data mining activities involving text processing and analysis, particularly in social media platforms where users employ a wide variety of abbreviations, slang, misspellings, and grammatical errors. To address this challenge, this study aimed to develop an algorithm for normalizing Malay noisy text using Levenshtein Distance (LD) and rule-based techniques. The LD is used to transform Malay spelling error words into their standard form, while rule-based techniques enhanced the conversion success rate for three categories of noisy term, namely slang, common Malay noisy text, and mixed language. The project was implemented using Python programming language, which demonstrated the effectiveness of the LD and rule-based techniques in normalizing noisy text in social media. The approach successfully normalized 80% of Malay noisy text into their standard text, which provides strong foundation for further study. Furthermore, this work open opportunities for introducing new approaches and rules to improve the normalization success rate, which can facilitate the analysis of text data in social media platforms. It is recommended that future studies focus on expanding the dataset and applying statistical validation methods to ensure the robustness and accuracy of the normalization model.

This article identifies the need to reduce the scale of misinformation and unethical business practices applied with the help of the use of artificial intelligence technology in online social media, among others, as an important element in managing the risk of developing applications and information systems equipped with artificial intelligence technology. The development of artificial intelligence based on big data and significant investments plays a key role in the economy and everyday life, automating processes, personalizing services and supporting sustainable development in fields such as medicine, energy and marketing. At the same time, it brings challenges, such as threats to jobs, privacy and ethics, which require appropriate regulation, development of technological infrastructure and public education. Unethical use of AI, such as algorithm manipulation in social media, amplifies misinformation and polarization, even though AI significantly improves efficiency and personalization of services. The issue of introducing the aforementioned regulations is an important element in reducing the scale of disinformation and unethical business practices used with the help of the use of AI technology in online social media, among others, and at the same time is an important element in managing the risk of developing applications and information systems equipped with AI technology.

This research aims to analyze the use of various languages ​​by initial and final level students at UPN Veteran East Java on social media X and its effects on social interaction. With the development of communication technology, social media has become the main platform for students to interact, share information and express themselves. The use of language on social media is influenced by many factors, including age, environment and level of education. In this context, freshmen and final year students have different backgrounds and experiences, which influence the way they communicate on digital platforms. The research method used is descriptive qualitative with data collection techniques in the form of observation. Data were analyzed to identify differences and similarities in the use of various languages ​​between the two groups of students. It is hoped that the results of this research will provide a deeper understanding of language dynamics among students, as well as the factors that influence their language choices on social media. These findings can also be a basis for developing better digital literacy among students and provide important insights into the dynamics of digital communication among the younger generation and their implications for the development of the Indonesian language in the modern era.

BackgroundSocial media provides dementia prevention researchers with additional opportunities to engage diverse audiences, including healthy individuals who unaware of their eligibility to take part in dementia‐related research. However, practical social media guidance that reflects the values and priorities of potential participants is needed. To address this gap, we sought to create consensus recommendations with research professionals and community experts.MethodWe conducted a three‐round, modified Delphi consisting of three online surveys and three conferences calls. Based on data from earlier project phases, a group of 16 experts with lived (n = 10) and professional (n = 6) experiences co‐created a set of recommendations to guide ethical social media use for dementia prevention research. Consensus was defined a priori as ≥70% agreement.ResultTwenty‐six items attained panelist agreement. Two privacy‐related items reached consensus in Round 1: the ethical appropriateness of closed social media groups (88%) and accessing individuals not on social media through social media contacts (79%). Nine items reached consensus in Round 2, including addressing misinformation (79%), stigma (93%), and other pertinent items for social media communication (e.g., public criticism, explaining process of science). Fifteen of the sixteen remaining items reached consensus after revision in Round 3. These items included defining appropriate comments (100%), rules of engagement on dementia social media pages (100%, e.g., positive messaging, relevant topics), and ranking appropriate prevention language use for different audiences (e.g., young, healthy adults, individuals with a family history). One item pertaining to language use for people living with dementia did not reach consensus.Recommendations were organized into seven social media use cases: setting up a social media page, handling online misinformation, actively challenging stigma, handling difficult online interactions, introducing new research to the public, help with study recruitment, and the language of prevention when writing posts.ConclusionResearch professionals and community members co‐created consensus recommendations to facilitate the ethical use of social media by dementia prevention researchers. By providing practical guidance, these recommendations will uphold ethical decision‐making on social media. Next steps are to create an evaluation tool and distribute the recommendations to relevant audiences.

Malay social media text is a text written on social media networks like Twitter. Commonly, this text comprises non-standard words, filled with dialects, foreign languages, word abbreviations, grammatical neglect, spelling errors, and many more. It is well known that this type of text is difficult to process due to its high noise and distinct text structure. Such problems can be resolved using rigorous text normalization, which is critical before any technique can be implemented and evaluated on social media text. In this paper, an improved normalization method towards Malay social media text was proposed by converting non-standard Malay words using a rule-based model. The method normalizes common language words often used by Malaysian users, such as non-standard Malay (like dialect and slangs), Romanized Arabic, and English words. Thus, a Malay text normalizer was proposed using a set of rules that extend across different domains of natural language processing (NLP) and is expected to address the challenges of processing Malay social media text. This study implements the proposed Malay text normalizer in a Part-of-Speech (POS) tagging application to evaluate the normalizer’s performance. The implementation demonstrates a substantial improvement in the POS tagging efficiency over several pre-processing stages, with an improvement of accuracy up to 31.8%. The increase of accuracy in the POS tagging indicates two main points. First, the Malay text normalizer’s rules improve the performance of a Malay text normalizer on social media text. Second, our proposed Malay text normalizer has successfully improved the POS tagging percentage and demonstrates the importance of normalized pre-processing in any NLP application.

Text normalization and sanitization are intrinsic components of Natural Language Inferences. In Information Retrieval or Dialogue Generation, normalization of user queries or utterances enhances linguistic understanding by translating non-canonical text to its canonical form, on which many state-of-the-art language models are trained. On the other hand, text sanitization removes sensitive information to guarantee user privacy and anonymity. Existing approaches to normalization and sanitization mainly rely on hand-crafted heuristics and syntactic features of individual tokens while disregarding the linguistic context. Moreover, such context-unaware solutions cannot dynamically determine whether out-of-vocab tokens are misspelt or are entity names. In this work, we formulate text normalization and sanitization as a multi-task text generation approach and propose a neural hybrid pointer-generator network based on multi-head attention. Its generator effectively captures linguistic context during normalization and sanitization while its pointer dynamically preserves the entities that are generally missing in the vocabulary. Experiments show that our generation approach outperforms both token-based text normalization and sanitization, while the hybrid pointer-generator improves the generator-only baseline in terms of BLEU4 score, and classical attentional pointer networks in terms of pointing accuracy.

One of the most persistent characteristics of written user-generated content (UGC) is the use of non-standard words. This characteristic contributes to an increased difficulty to automatically process and analyze UGC. Text normalization is the task of transforming lexical variants to their canonical forms and is often used as a pre-processing step for conventional NLP tasks in order to overcome the performance drop that NLP systems experience when applied to UGC. In this work, we follow a Neural Machine Translation approach to text normalization. To train such an encoder-decoder model, large parallel training corpora of sentence pairs are required. However, obtaining large data sets with UGC and their normalized version is not trivial, especially for languages other than English. In this paper, we explore how to overcome this data bottleneck for Dutch, a low-resource language. We start off with a small publicly available parallel Dutch data set comprising three UGC genres and compare two different approaches. The first is to manually normalize and add training data, a money and time-consuming task. The second approach is a set of data augmentation techniques which increase data size by converting existing resources into synthesized non-standard forms. Our results reveal that, while the different approaches yield similar results regarding the normalization issues in the test set, they also introduce a large amount of over-normalizations.

Objective: To assess the frequency of antibiotic use through internet and social media among the people visiting Rawal institute of health sciences. To assess knowledge about antibiotics and to explore role of internet search engines and social media for accessing information regarding antibiotic use. Study Design: Cross-sectional Descriptive study. Setting: Rawal Institute of health sciences, Islamabad, Pakistan. Period: June 2019 to September 2019. Material & Methods: A was study carried to find out the role of internet and social media for use of antibiotics through a structured questionnaire from 150 people visiting RIHS. Sampling was done through non probability convenient sampling method. Response rate was 100%. Results: The data of 150 participants showed 92% have used antibiotics in their life time, of those 92% (n=138) participants, 26 %( n=39) have used them through prescription, 30% (n=45) have done self-medication, 19% (n=29) have used social media for medication. 91% think that information regarding use of antibiotics can be spread through internet and social media. Conclusion: The results conclude that social media and internet related antibiotic sales is contributing a substantial share. Majority of the participant also think that internet and social media are the best source to spread information for safe use. National guidelines should be devised for appropriate information seeking and communication interventions can be done using internet and social media.