Large User Research Articles

Responsible Crowdsourcing for Responsible Generative AI: Engaging Crowds in AI Auditing and Evaluation

With the rise of generative AI (GenAI), there has been an increased need for participation by large and diverse user bases in AI evaluation and auditing. GenAI developers are increasingly adopting crowdsourcing approaches to test and audit their AI products and services. However, it remains an open question how to design and deploy responsible and effective crowdsourcing pipelines for AI auditing and evaluation. This workshop aims to take a step towards bridging this gap. Our interdisciplinary team of organizers will work with workshop participants to explore several key questions, such as how to improve the output quality and workers' productivity for GenAI evaluation crowdsourcing tasks compared to discriminative AI systems, how to guide crowds in auditing problematic AI-generated content while managing their psychological impact, ensuring marginalized voices are heard, and setting up responsible and effective crowdsourcing pipelines for real-world GenAI evaluation. We hope this workshop will produce a research agenda and best practices for designing responsible crowd-based approaches to AI auditing and evaluation.

A Beam Hopping Scheme Based on Adaptive Beam Radius for LEO Satellites.

Toward the vision of seamless global connectivity in the 6G era, the non-terrestrial network (NTN) in space-air-ground integrated networks (SAGINs) network architecture is one of the highly promising solutions. From the perspective of relay nodes, NTN includes satellite nodes and space-based platform nodes. As a resource management technology in satellite communication, beam-hopping has garnered significant attention from researchers due to its effectiveness in ad-dressing the disparity between offered capacities and uneven terrestrial traffic demands. Recognizing that the larger beams offer broader coverage but the smaller ones provide better an-ti-interference capabilities and higher throughput, this paper introduces an adaptive cluster-ing-based approach. It provides large, medium, and small user beams to target ground users. The proposed algorithm aims to minimize total system latency and enhance system throughput. Sim-ulation results show that employing the proposed algorithm in the baseline model results in a 3.44% increase in system throughput and a 35.5% reduction in system latency. Furthermore, simulation results based on alternative models indicate that while the proposed algorithm may lead to a slight decrease in system throughput, it brings significant improvements in system latency.

Reinventing BrED: A Practical Construction

Broadcast Encryption (BE) allows a sender to send an encrypted message to multiple receivers. In a typical broadcast encryption scenario, the broadcaster decides the set of users who can decrypt a particular ciphertext (denoted as the privileged set). Gritti et al. (IJIS'16) introduced a new primitive called Broadcast Encryption with Dealership (BrED), where the dealer decides the privileged set. A BrED scheme allows a dealer to buy content from the broadcaster and sell it to users. It provides better flexibility in managing a large user base. To date, quite a few different constructions of BrED schemes have been proposed by the research community. We find that all existing BrED schemes are insecure under the existing security definitions. We demonstrate a concrete attack on all the existing schemes in the purview of the existing security definition. We also find that the security definitions proposed in the state-of-the-art BrED schemes do not capture the real world. We argue about the inadequacy of existing definitions and propose a new security definition that models the real world more closely. Finally, we propose a new BrED construction and prove it to be secure in our newly proposed security model.

Analysis of Profiles of Supporters of Conspiracy Narratives about Vaccination Against COVID-19 on a Social Network

The article analyzes the socio-demographic characteristics and value-worldview self-positioning of the distributors of conspiracy narratives (“conspiracy theories”) about vaccination against COVID-19 on the Russian social network “VK”. The research material consists of 33,674 profiles of VK users who published at least two messages during the pandemic that explicate the stated narrative. The collection and analysis of the material was carried out using big data analysis technologies. The purpose of the study is to determine whether it is possible to identify typical socio-demographic and ideological traits of anti-vaccer conspiracy theorists, which to a certain extent explain their penchant for conspiracy thinking. The results obtained refute the marginal dominance of conspiracy theorists in a pandemic situation. Versions popular in the scientific literature about the predominance of insufficiently educated people among “conspiracy theorists”, or the elderly as more gullible, or women as more suggestible, or residents of villages and small provincial towns as more conservative, etc. are not confirmed by the large-scale empirical studies of large user data. The results of the study may be useful for developing countermeasures and strategies to address public mistrust of institutions and official trends in public health, in particular vaccination policies. The traditional focus on information and educational measures (explanations about the dangers of the disease and the benefits of the vaccine) seems to be ineffective in the context of the emergence of value systems, cognitive distortions, general socio-political tension and other reasons that are still awaiting serious study.

Researching hate speech online: Exploring the potential and limitations of Facebook as a survey tool in Africa

The reasons driving users to flag/report online hate and social media mis/disinformation in Africa—or, conversely, why they choose not to. We conducted surveys with social media users in several African countries to gauge their definitional understandings of online hate and mis/disinformation; their awareness of content moderation processes, including the role of artificial intelligence (AI); and their engagement with flagging/reporting online content. The challenge, however, was how to conduct survey research in difficult-to-reach places, some of which are affected by violent conflict, and within a limited budget. Given that our aim was to reach social media users, we used Facebook to advertise our survey, drawing on the breadth of research on using the platform as a research tool and the fact it has a large user base in the targeted countries. In this article, we explore the practicalities, challenges, and limitations of Facebook as a tool for survey research in African countries.

Pengembangan Website Desa Wisata Sebagai Media Informasi Wisatawan Pada Desa Temesi

The tourism industry is undergoing significant changes with the advancement of digital technology, known as digital tourism. Digitalization has changed the way tourist destinations are promoted, accessed, and managed, including through the concept of rural tourism or tourism villages. Temesi Village in Gianyar Regency, Bali, has great natural and cultural tourism potential. However, to optimize this potential, digital media is needed, such as website creation. The WordPress platform was chosen for its advantages, including a large user community, ease of use, design flexibility, and data security. This step aligns with the digital village program supported by the Community Empowerment Agency and Primakara University. It is hoped that the creation of this website will introduce and promote the tourism potential of Temesi Village to the general public and tourists, as well as support sustainable tourism development in the area.

A User-Centered Framework for Data Privacy Protection Using Large Language Models and Attention Mechanisms

This paper introduces a user-centered data privacy protection framework utilizing large language models (LLMs) and user attention mechanisms, which are tailored to address urgent privacy concerns in sensitive data processing domains like financial computing and facial recognition. The innovation lies in a novel user attention mechanism that dynamically adjusts attention weights based on data characteristics and user privacy needs, enhancing the ability to identify and protect sensitive information effectively. Significant methodological advancements differentiate our approach from existing techniques by incorporating user-specific attention into traditional LLMs, ensuring both data accuracy and privacy. We succinctly highlight the enhanced performance of this framework through a selective presentation of experimental results across various applications. Notably, in computer vision, the application of our user attention mechanism led to improved metrics over traditional multi-head and self-attention methods: FasterRCNN models achieved precision, recall, and accuracy rates of 0.82, 0.79, and 0.80, respectively. Similar enhancements were observed with SSD, YOLO, and EfficientDet models with notable increases in all performance metrics. In natural language processing tasks, our framework significantly boosted the performance of models like Transformer, BERT, CLIP, BLIP, and BLIP2, demonstrating the framework’s adaptability and effectiveness. These streamlined results underscore the practical impact and the technological advancement of our proposed framework, confirming its superiority in enhancing privacy protection without compromising on data processing efficacy.

Impact of Short Video Marketing on Film Promotion: A Case Study of Douyin Platform

Short video platforms, with their ease of content production, wide dissemination, and large user base, have become important mediums in the internet, providing new opportunities for film marketing and promotion. This study explores the use and effects of short video platforms (Douyin) in film marketing. Through case analysis and in-depth interviews, short video marketing has significant advantages in disseminating film information, attracting audiences with diverse content forms, and increasing exposure and appeal. However, issues like excessive plot leakage and content homogenization may affect audience expectations and viewing experience. The study also reveals that short video marketing excels in emotional resonance and guiding social topics but should avoid over-reliance on star power and misleading promotions. Based on these findings, the study recommends optimizing short video film marketing strategies, including maintaining content novelty, enhancing emotional marketing, effectively using audiovisual language, and accurately targeting audiences to improve marketing effectiveness and audience engagement.

Techno-Economic Assessment of Electricity Generation From a Medium-Scale CSP-PV Hybrid Plant Using Long-Duration Storage

The South African electricity sector is transitioning from a single utility buyer model to an open market model due to increasing utility tariffs, the energy crisis, sustainability goals, and enabling legislation. As a result, the private sector is seeking to procure private power generation technologies, with solar photovoltaics (PV) and onshore wind being the primary choices. However, the integration of these technologies is limited due to their variability and lack of dispatchability. Concentrating solar power (CSP) tower technology combined with long-duration storage is gaining traction in high solar resource regions due to efficiency improvements and cost reductions. This paper evaluates a medium-scale CSP-PV hybrid plant’s technical and economic feasibility in South Africa, focusing on cost savings, energy security, and sustainability. The study identifies a hybrid plant suitable for large power users, with independent power producers (IPPs) developing such a plant using non-recourse project finance and energy sold to large users through long-term power purchase agreements (PPA). The first-year PPA tariffs are determined using a typical project finance model developed for local market conditions. Depending on the plant design and the off-taker’s utility tariff structure, the plant offers cost savings (and average tariffs) ranging from 9.25% (at 78 USD/MWh) to 17.58% (at 81 USD/MWh). Hybrid plants are expected to become more feasible and “bankable” with improving technology and decreasing costs in South Africa.

Research on the Development Strategy of Chinese Second-Dimensional Mobile Games - A Case Study Based on Genshin Impact

With the rapid development of technology and the continuous growth of people's entertainment needs, second-dimensional mobile games have become an undeniable force in the global gaming market, with limitless prospects and business opportunities. However, to stand out in this fiercely competitive market, it is still necessary to carefully explore the development strategies and make judgments and decisions that fit the times. This paper takes the development of Genshin Impact, a leading Chinese second-dimensional game, as a specific case, and uses the SWOT analysis method to explore its advantages, disadvantages, opportunities, and threats in development, thereby studying the current state of the second-dimensional mobile gaming industry. The research results show that Genshin Impact has the advantages of a large second-dimensional user base with strong spending power and significant resource investment by its developer, miHoYo; but also faces the disadvantage of a significant reduction in player numbers; at the same time, it encounters opportunities with the increase in residents' income and the rapid development of cloud gaming platforms; as well as threats from intense market competition and strict game regulatory policies in China. Furthermore, based on the results of the SWOT analysis, this paper provides suggestions for strengthening innovation and cooperation, increasing game publicity efforts, improving regulatory response capabilities, and expanding revenue channels, pointing the way for the future development of Genshin Impact and the second-dimensional mobile gaming industry.

IDify - Distributed Database System for Digital Identification

”IDify” Distributed database system for digital identification is a big data handling system to centralize the important services provided by the government to make the facilities efficient and reliable. It has integrated services such as digital authentication, E-Banking, Personal Information Management, and Government Surveys to provide users with a centralized and digitized platform to make their lives easier. In this system there is a distributed database platform to carry out all the data storing and data processing tasks to facilitate a large user count and to avoid the drawbacks of monolithic systems. The database system has been designed by integrating several modern database technologies to make the system efficient and coherent as the facilities provided by the system are extremely mission-critical. To make the system low latent on data ingestion and data reading tasks a special library is also implemented in collaboration with modern query processing engines. It is proposed to utilize the thread system of a processor to implement parallelism to enhance the performance of data read/write tasks. Apart from that a fault-tolerant framework has been implemented to free the system from any erroneous scenario taking place.

An effective attention and residual network for malware detection

AbstractDue to its open source and large user base, Android has emerged as the most popular operating system. Android's popularity and openness have made it a prime target for malicious attackers. Permissions have received great attention from researchers because of their effectiveness in restricting applications’ access to sensitive resources. However, existing malware detection methods based on permissions are easily bypassed by inter‐application resource access. To address these issues, we combine inter‐application resource access‐related intent features with permission features. Besides, we designed a customized convolutional neural network using two squeeze‐and‐excitation blocks to learn the inherent relationships between multi‐type features. The two basic SE blocks perform squeezing operations based on average pooling and max pooling, respectively, to compute channel‐wise attention from multiple perspectives. We designed a series of experiments based on real‐world samples to evaluate the efficacy of the proposed framework. Empirical results demonstrate that our framework outperforms state‐of‐the‐art methods, achieving an accuracy of 96.29%, precision of 97.52%, recall of 94.63%, F1‐score of 96.06% and MCC of 92.60%. These promising experimental results consistently demonstrate that AMERDroid is an effective approach for Android malware detection.

Large Language Models and User Trust: Consequence of Self-Referential Learning Loop and the Deskilling of Health Care Professionals.

As the health care industry increasingly embraces large language models (LLMs), understanding the consequence of this integration becomes crucial for maximizing benefits while mitigating potential pitfalls. This paper explores the evolving relationship among clinician trust in LLMs, the transition of data sources from predominantly human-generated to artificial intelligence (AI)-generated content, and the subsequent impact on the performance of LLMs and clinician competence. One of the primary concerns identified in this paper is the LLMs' self-referential learning loops, where AI-generated content feeds into the learning algorithms, threatening the diversity of the data pool, potentially entrenching biases, and reducing the efficacy of LLMs. While theoretical at this stage, this feedback loop poses a significant challenge as the integration of LLMs in health care deepens, emphasizing the need for proactive dialogue and strategic measures to ensure the safe and effective use of LLM technology. Another key takeaway from our investigation is the role of user expertise and the necessity for a discerning approach to trusting and validating LLM outputs. The paper highlights how expert users, particularly clinicians, can leverage LLMs to enhance productivity by off-loading routine tasks while maintaining a critical oversight to identify and correct potential inaccuracies in AI-generated content. This balance of trust and skepticism is vital for ensuring that LLMs augment rather than undermine the quality of patient care. We also discuss the risks associated with the deskilling of health care professionals. Frequent reliance on LLMs for critical tasks could result in a decline in health care providers' diagnostic and thinking skills, particularly affecting the training and development of future professionals. The legal and ethical considerations surrounding the deployment of LLMs in health care are also examined. We discuss the medicolegal challenges, including liability in cases of erroneous diagnoses or treatment advice generated by LLMs. The paper references recent legislative efforts, such as The Algorithmic Accountability Act of 2023, as crucial steps toward establishing a framework for the ethical and responsible use of AI-based technologies in health care. In conclusion, this paper advocates for a strategic approach to integrating LLMs into health care. By emphasizing the importance of maintaining clinician expertise, fostering critical engagement with LLM outputs, and navigating the legal and ethical landscape, we can ensure that LLMs serve as valuable tools in enhancing patient care and supporting health care professionals. This approach addresses the immediate challenges posed by integrating LLMs and sets a foundation for their maintainable and responsible use in the future.

Quality of Answers of Generative Large Language Models Versus Peer Users for Interpreting Laboratory Test Results for Lay Patients: Evaluation Study.

Although patients have easy access to their electronic health records and laboratory test result data through patient portals, laboratory test results are often confusing and hard to understand. Many patients turn to web-based forums or question-and-answer (Q&A) sites to seek advice from their peers. The quality of answers from social Q&A sites on health-related questions varies significantly, and not all responses are accurate or reliable. Large language models (LLMs) such as ChatGPT have opened a promising avenue for patients to have their questions answered. We aimed to assess the feasibility of using LLMs to generate relevant, accurate, helpful, and unharmful responses to laboratory test-related questions asked by patients and identify potential issues that can be mitigated using augmentation approaches. We collected laboratory test result-related Q&A data from Yahoo! Answers and selected 53 Q&A pairs for this study. Using the LangChain framework and ChatGPT web portal, we generated responses to the 53 questions from 5 LLMs: GPT-4, GPT-3.5, LLaMA 2, MedAlpaca, and ORCA_mini. We assessed the similarity of their answers using standard Q&A similarity-based evaluation metrics, including Recall-Oriented Understudy for Gisting Evaluation, Bilingual Evaluation Understudy, Metric for Evaluation of Translation With Explicit Ordering, and Bidirectional Encoder Representations from Transformers Score. We used an LLM-based evaluator to judge whether a target model had higher quality in terms of relevance, correctness, helpfulness, and safety than the baseline model. We performed a manual evaluation with medical experts for all the responses to 7 selected questions on the same 4 aspects. Regarding the similarity of the responses from 4 LLMs; the GPT-4 output was used as the reference answer, the responses from GPT-3.5 were the most similar, followed by those from LLaMA 2, ORCA_mini, and MedAlpaca. Human answers from Yahoo data were scored the lowest and, thus, as the least similar to GPT-4-generated answers. The results of the win rate and medical expert evaluation both showed that GPT-4's responses achieved better scores than all the other LLM responses and human responses on all 4 aspects (relevance, correctness, helpfulness, and safety). LLM responses occasionally also suffered from lack of interpretation in one's medical context, incorrect statements, and lack of references. By evaluating LLMs in generating responses to patients' laboratory test result-related questions, we found that, compared to other 4 LLMs and human answers from a Q&A website, GPT-4's responses were more accurate, helpful, relevant, and safer. There were cases in which GPT-4 responses were inaccurate and not individualized. We identified a number of ways to improve the quality of LLM responses, including prompt engineering, prompt augmentation, retrieval-augmented generation, and response evaluation.

Detecting multilingual hate speech targeting immigrants and women on Twitter

Twitter has experienced a tremendous surge in popularity over recent years, establishing itself as a prominent social media platform with a large user base. However, with this increased usage, there has been a concerning rise in the number of individuals resorting to derogatory language and expressing their opinions in a demeaning manner toward others. This surge in hate speech has drawn significant attention to the field of sentiment analysis, which aims to develop algorithms capable of detecting and analyzing emotions expressed in social networks using intuitive approaches. This paper focuses on addressing the complex task of detecting hate speech and aggressive behavior while performing target classification. We explored various deep-learning approaches, including LSTM, BiLSTM, CNN, and GRU. Each offers unique capabilities for capturing different aspects of the input data. We proposed an ensemble approach that combines the top three performing models. This ensemble approach benefits from the diverse strengths of each individual model showing F1 score of 0.85 for English-HS, 0.94 for English-TR, 0.92 for English-AB, 0.84 for Spanish-HS, 0.86 for Spanish-TR, 0.97 for Spanish-AB, 0.74 for multilingual-HS, 0.94 for multilingual-TR, and 0.88 for multilingual-AB.

Unveiling the Synergistic Relationship between Distributed Systems and Cloud Computing: A Review of Architectural Trends

Cloud providers use distributed systems for scalability, availability, performance, automation, multi-tenancy, and innovation. Distributed cloud computing distributes workload across multiple locations, improving application performance and responsiveness. Significantly potential computational resources are developed in cloud, where large-scale, intricate tasks are performed with the backbone of distribute infrastructure in cloud systems, similar to supercomputing. Cloud computing development has significantly impacted software development and testing, necessitating applications compatible with the cloud, large data users, and high security. Distributed applications hoist on to cloud platforms where increased efficiency, reliability and low costs are favored and further be stored in the cloud for flexibility and scalability. Cloud service models include Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS), each offering different application services, programming languages, and hosting environments. The synergistic aspects of Distributed Systems and Cloud Systems with respect to their basic capabilities are discussed and systematically reviewed.

A Transfer Learning-based Method for the Daily Electricity Consumption Forecasting of Large Industrial Users after Business Expansion

Background: With the rapid development of industry, the expansion capacity and frequency of large industrial users continue to increase. However, the traditional static prediction model is difficult to accurately predict the daily electricity consumption of industrial expansion, which is not conducive to the safe and stable operation of the power grid. Objective: In response to the above problems, this paper proposes a transfer learning-based method for the daily electricity consumption forecasting of large industrial users after business expansion. Methods: Firstly, a dynamic training framework for the prediction model of transfer learning is established, so that the prediction model can dynamically adapt to the capacity change brought about by the expansion of multi-user business. Then, a neural network for predicting daily electricity consumption of industrial users based on multi-resolution time series attention is established, which can deeply mine the characteristics of electricity sequence. Finally, a deep learning model parameter migration and adjustment method considering business expansion is proposed, which can realize efficient migration of prediction models. Results: The effectiveness of the proposed method is demonstrated by comparing it with state-ofthe- art electricity forecasting based on two-year historical data of a specific region. Conclusion: The proposed method is compared with state-of-the-art power forecasting techniques through the validation of local historical data. The obtained results demonstrate the effectiveness of the proposed method.

An Analysis of Households Choice of Solid Fuels as a Primary and Supplementary Heating Fuel

The residential sector in Ireland is a large user of solid fuels for space heating purposes. Solid fuels are commonly used to supplement other forms of heating rather than as the primary source. Using a survey data set of Irish households and a multinomial logit approach, differences between the household characteristics of primary and supplementary solid fuel users are identified, including for levels of education, age of dwelling, location and pro-environmental attitudes. Evidence also shows that increases in income lead to a transition away from primary solid fuel use but not supplementary consumption, suggesting that an energy stacking model explains the household’s choice of heating fuels in Ireland. Given the established effects that solid fuels have on air quality and the scale of supplementary solid fuel use, policies to promote a transition to cleaner fuels need to account for the clear differences in the features of the two user groups.

Analysis of Anonymous Student Narratives About Experiences with Emergency Medicine Residency Programs.

Academic emergency medicine (EM) communities have viewed anonymous online communities (AOC) such as Reddit or specialty-specific "applicant spreadsheets" as poor advising resources. Despite this, robust EM AOCs exist, with large user bases and heavy readership. Insights about applicants' authentic experiences can be critical for applicants and program leadership decision-making. To date, there are no EM studies to qualitatively assess EM AOC narratives during the application cycle. Our goal was to perform a qualitative analysis of students' EM program experiences through a publicly available AOC. This was a qualitative analysis of a publicly available, time-stamped, user-locked AOC dataset: "Official 2020-2021 Emergency Medicine Applicant Spreadsheet." We extracted and then de-identified all data from selected sub-sheets entitled "Virtual Interview Impressions" and "Rotation Impressions." Four investigators used constant comparative method to analyze the data inductively, and they subsequently met to generate common themes discussed by students. Preliminary thematic analysis was conducted on a random sample of 37/183 (20%) independent narratives to create the initial codebook. This was used and updated iteratively to analyze the entire narrative set consisting of 841 discrete statements. Finally, two unique codes were created to distinguish whether the identified sub-themes, or program attributes, were likely "modifiable" or "non-modifiable." We identified six major themes: living and working conditions; interpersonal relationships; learning experiences, postgraduate readiness, and online/virtual supplements. Common sub-themes included patient population (13%); resident personality (7%); program leadership personality (7%); relationship with faculty/leadership (6%); geography (4%); practice setting (4%); program reputation (4%), and postgraduate year-3 experiences (4%). Modifiable sub-themes outnumbered non-modifiable sub-themes, 60.7% to 39.3%. In this analysis of selected medical students' narratives in an AOC, the majority of identified themes represented topics that may serve as external feedback for EM residency programs and their clerkships. Selective use of AOCs may set a precedent for future program assessments by applicants and inform program leadership of important programmatic elements in the eyes of applicants. It elucidates important themes in their interactions or learning experiences with programs and creates opportunities for learner-centric program improvement.

Fake News Identification in Urdu Tweets Using Machine Learning Models

There is an increasing number of people who generate and distribute content online, especially via social media platforms, which is primarily responsible for the proliferation of fake information. Fake information can cause controversy and distort people's perspectives, so it needs to be addressed immediately. The goal of this work is to detect false information in Urdu tweets, a difficult task given the language's large user population and particular grammatical difficulties. We offer an all-inclusive machine learning system that reliably classifies tweets in Urdu as legitimate or false. The methodology we use consists of several key steps: preprocessing, which includes normalizing, tokenizing, removing stop words, and stemming to prepare the data for analysis; data collection, which involves compiling and annotating a sizable dataset of Urdu tweets; and feature extraction, which makes use of technique TF-IDF to extract the semantic and syntactic nuances of the language. We investigate various machine learning models, including RNNs and CNNs, and more sophisticated neural networks like SVM, Random Forest, Logistic Regression, Naive Bayes, and Decision Tree to find the most efficient method for resolving this classification problem. The models are put through a rigorous training and assessment process using measures including the F1 score, accuracy, precision, and recall. Furthermore, a thorough examination of their confusion matrices is done. Our study's findings suggest that deep learning models hold much promise for resolving the problem of inaccurate information in Urdu. This opens the door for additional research and the creation of real-time algorithms for spotting false information. The subject of information integrity in Urdu language content is improved by this work, which also sheds light on the applicability of machine learning techniques in many linguistic contexts. Using SVM, Random Forest, Logistic Regression, Naive Bayes, and Decision Tree we achieved accuracies of 74%,91%, 76%, 78%, and 67% respectively. Meanwhile, CNN and RNN are the classifiers with the highest accuracy levels at 91% and 99% respectively. The results demonstrate that the CNN Model achieved 99% highest accuracy in detecting fake news from Urdu Tweets.