Search Results (50)

Advanced Driver Assistance Systems (ADAS) and Automated Driving Systems (ADS) are becoming convenient modes of transportation; however, their safety remains a critical concern as crashes continue to occur. To reveal crash trends and temporal variations, this study develops time-series forecasting models to predict future crash counts of such vehicles. The crash dataset released by the National Highway Traffic Safety Administration (NHTSA) has been used here. Two univariate forecasting models—the Seasonal Autoregressive Integrated Moving Average (SARIMA) and the Facebook Prophet model—have been used here for different datasets. The models were trained on 30 months of data (July 2021 to December 2023) and validated on 6 months of data (January–June 2024). Validation metrics include Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Theil’s U1 statistic. Results showed that Facebook Prophet significantly outperformed SARIMA for both datasets, achieving an RMSE of 2.71 and an MAPE of 6.9% for ADAS, and an RMSE of 2.24 and an MAPE of 8.85% for ADS. For both systems, the model revealed empirically observed cyclical patterns and consistent rising trends. ADAS crashes exhibit a bimodal temporal pattern, with recurring peaks in January and May–June, alongside notable troughs in February–March and August–September. ADS displays a trimodal pattern, with recurring peaks in April–May, August and October, alongside notable troughs in December and the early winter months. These patterns represent empirically identified temporal regularities rather than causally attributed seasonality. From the future forecasts for July to December 2024, the model showed that ADAS crashes are expected to range between 40 and 80 per month, while ADS crashes are projected to remain between 20 and 40 per month. These findings underscore the need for proactive safety measures and enhanced regulatory oversight during identified high-risk periods to mitigate the growing trend in AV crashes. Full article

Understanding how moral and emotional language operates in paid social advertising is essential for evaluating persuasion and its ethical contours. We provide a descriptive map of Moral Foundations Theory (MFT) language in Meta ad copy (Facebook/Instagram) drawn from seven global beverage brands across eight English-speaking markets. Using the moralstrength toolkit, we implement a two-channel pipeline that combines an unsupervised semantic estimator (SIMON) with supervised classifiers, enforces a strict cross-channel consensus rule, and adds a non-overriding purity diagnostic to reduce attribute-based false positives. The corpus comprises 758 text units, of which only 25 ads (3.3%) exhibit strong consensus, indicating that much of the copy is either non-moral or linguistically ambiguous. Within this high-consensus subset, the distribution of moral cues varies systematically by brand and category, with loyalty, fairness, and purity emerging as the most prominent frames. A valence pass (VADER) indicates that moralized copy tends toward negative valence, yet it may still yield a constructive overall tone when advertisers follow a crisis–resolution structure in which high-intensity moral cues set the stakes while surrounding copy positions the brand as the solution. We caution that text-only models undercapture multimodal signaling and that platform policies and algorithmic recombination shape which moral cues appear in copy. Overall, the study demonstrates both the promise and the limits of current text-based MFT estimators for advertising: they support transparent, reproducible mapping of moral rhetoric, but future progress requires multimodal, domain-sensitive pipelines, policy-aware sampling, and (where available) impression/spend weighting to contextualize descriptive labels. Full article

Advertising research highlights the crucial role of creative strategy in shaping consumer behavior. Yet, limited attention has been paid to how creative appeal and message strategy jointly influence persuasion in social media contexts. This study examines the interactive effects of informational versus transformational appeals and personal versus social-experience message strategies on consumer attitudes and purchase intentions. A 2 (creative appeal) × 2 (message strategy) experimental design was implemented using Facebook post advertisements for a fictitious beer brand. Data was collected from 231 participants randomly assigned to one of four ad conditions. Results show that informational appeals outperform transformational appeals in generating immediate purchase intentions. Attitudes toward the ad and attitude toward the brand mediated these effects, consistent with the Dual Mediation Hypothesis. Moreover, in accordance with the Construal Level Theory, message strategy moderates the relationship: informational appeals were most effective when paired with personal strategies but lost persuasive power under social-experience strategies. These findings advance the theoretical understanding of digital advertising persuasion by explicating how creative appeal and message strategy jointly shape both attitudinal and behavioral responses. Practically, the results suggest that advertisers seeking short-term conversions should combine informational appeals with personal strategies. Full article

Hate speech is a complex topic that transcends language, culture, and even social spheres. Recently, the spread of hate speech on social media sites like Facebook has added a new layer of complexity to the issue of online safety and content moderation. This study seeks to minimize this problem by developing an Arabic script-based tool for automatically detecting hate speech in Roman Urdu, an informal script used most commonly for South Asian digital communications. Roman Urdu is relatively complex as there are no standardized spellings, leading to syntactic variations, which increases the difficulty of hate speech detection. To tackle this problem, we adopt a holistic strategy using a combination of six machine learning (ML) and four Deep Learning (DL) models, a dataset from Facebook comments, which was preprocessed (tokenization, stopwords removal, etc.), and text vectorization (TF-IDF, word embeddings). The ML algorithms used in this study are LR, SVM, RF, NB, KNN, and GBM. We also use deep learning architectures like CNN, RNN, LSTM, and GRU to increase the accuracy of the classification further. It is proven by the experimental results that deep learning models outperform the traditional ML approaches by a significant margin, with CNN and LSTM achieving accuracies of 95.1% and 96.2%, respectively. As far as we are aware, this is the first work that investigates QLoRA for fine-tuning large models for the task of offensive language detection in Roman Urdu. Full article

In 2022, 26–31% of young adults reported using e-cigarettes in the previous 30 days. Research supports the effectiveness of mass media health campaigns in changing targeted attitudes and behaviors regarding nicotine use. However, the effect of social media campaigns and the pathway through which they change attitudes and behaviors require more research. This randomized controlled experiment examines the pathway through which exposure to an e-cigarette prevention social media campaign influences intentions to quit e-cigarettes among young adults who currently use e-cigarettes. Participants (n = 160) aged 18 to 24 years old were recruited through Virtual Lab in Facebook and Instagram. Structural equation modeling (SEM) was used to examine the pathway from campaign exposure, to changes in targeted attitudes, and finally to intentions to quit e-cigarettes in the next year. Ad exposure was significantly associated with stronger anti-industry attitudes, independence from e-cigarettes, and risk perceptions. These campaign-targeted attitudes were significantly associated with greater intentions to quit e-cigarettes (anti-industry attitudes (OR = 1.43), independence (OR = 1.50), and risk perception (OR = 1.71)). The findings suggest that exposure to an e-cigarette prevention social media campaign can affect targeted attitudes, which in turn improve intentions to quit. Future research should examine behavior changes and compare the effects between those currently using e-cigarettes and those not using them. Full article

Background: Facebook groups have been used to foster social support of transplant patients. Examining the use and content strategies for generating member interactions within transplant-specific groups can inform how we leverage these groups to expand access to social support resources. This study characterizes the use of a closed Facebook group for kidney transplant patient support linked to a hospital in Buffalo, NY to identify the most engaging content. Methods: The sample consisted of 387 individuals (372 patients/family, eight transplant professionals, and seven community advocates) and the administrator. Content analysis was conducted of posts and comments made to the group. Descriptive measures of post content associated with interactions (reactions and comments) were computed. Results: Between 5/2020 and 5/2023, there were 484 posts with 8233 interactions (2793 comments, 5440 reactions). Half of the posts (n = 241) were made by the administrator, 166 (34%) by patients/family, 70 (14%) by community advocates, and 7 (1%) by transplant professionals. Content analysis revealed that post types with the most interactions were personal experiences, monthly transplant volume, and monthly new members added; the least interactions involved posts about holidays, observances, and information. Conclusions: The interaction metrics varied according to the content strategies used by members and provided insights into the types of content members interact with. Full article

The mass media are central to everyday life, a meeting point for citizens with the facts of the current social situation, and a space for the meaning, perception, interpretation, and construction of the notion of reality in the collective imagination. The impact of technology and communication platforms on the social fabric has opened up access to information and atomized trust and credibility in the face of the journalistic brand, an instance of crisis on the smartphone screen in which the media are relegated to the background and influencers, opinion leaders, and the protagonists of the facts themselves, in direct contact with the followers, gain relevance. The transformation configured in the users’ content diet aims to review the industry’s behavior on social networks to interpret why it is losing its place in the citizens’ consumption agenda. This research, which relied on a mixed methodology, downloaded messages from Facebook, X/Twitter, Instagram, YouTube, and TikTok profiles of two written, five audio, and two audiovisual media in a period of 91 days, and added them to what was carried out in the same period in 2019 and 2021. The corpus of the study made it possible to prove the composition of the communication, the operational tactics, and the thematic axis. The research concluded that the Colombian mass media need to innovate to establish productive routines that bet on digital native products, corresponding to the dynamics of networked consumption by an audience that concentrates all actions on screen devices. Full article

Spam communications from spam ads and social media platforms such as Facebook, Twitter, and Instagram are increasing, making spam detection more popular. Many languages are used for spam review identification, including Chinese, Urdu, Roman Urdu, English, Turkish, etc.; however, there are fewer high-quality datasets available for Urdu. This is mainly because Urdu is less extensively used on social media networks such as Twitter, making it harder to collect huge volumes of relevant data. This paper investigates policy-based Urdu tweet spam detection. This study aims to collect over 1,100,000 real-time tweets from multiple users. The dataset is carefully filtered to comply with Twitter’s 100-tweet-per-hour limit. For data collection, the snscrape library is utilized, which is equipped with an API for accessing various attributes such as username, URL, and tweet content. Then, a machine learning pipeline consisting of TF-IDF, Count Vectorizer, and the following machine learning classifiers: multinomial naïve Bayes, support vector classifier RBF, logical regression, and BERT, are developed. Based on Twitter policy standards, feature extraction is performed, and the dataset is separated into training and testing sets for spam analysis. Experimental results show that the logistic regression classifier has achieved the highest accuracy, with an F1-score of 0.70 and an accuracy of 99.55%. The findings of the study show the effectiveness of policy-based spam detection in Urdu tweets using machine learning and BERT layer models and contribute to the development of a robust Urdu language social media spam detection method. Full article

In this paper, we look at two research questions. First, can lower ad-valorem taxes, on the selling of news and on the selling of advertising, conduce to lower prices in the media sector? Second, can lower ad-valorem taxes stimulate firms to increase the diversity of content that they offer? The purpose of this work is to give tax political guidelines to policy makers for the media sector. This is important for a sector that has seen the reduction in payment subscriptions by readers (due to competition from free news from the Internet), and reduction of advertisement revenues due to competition from media giants like Google and Facebook. With this purpose we build on the Hoteling product competition model, which is the workhorse model in media economics. We show that ad-valorem taxes on the selling of advertising are preferable to ad-valorem taxes on the selling of news because the former conduce to reduction in prices of newspaper. However, both ad-valorem taxes on the selling of news and on the selling of advertisement reduces media diversity, because they reduce revenues that media firms can use to invest in media content. Full article

The thesis of this paper is that in the digital age we are moving away from words and concepts characteristic of the print age, towards the use of images and symbols. I distinguish between objective symbols as in mathematics, and cultural symbols as in poetry and religion. Students must learn to move from recognizing the objective rules of language to internalizing the norms of culture, according to the analogy of learning. Ricoeur’s theory of interpretation explains the passage from recognition to interpretation in the cultural sciences. This passage is not only cognitive but also implies the discovery of an experiential dimension, as in poetry and worship. This theory is applied to the findings from religious ads on Facebook. By creating new audiences by trial and error, the number of viewers increased from 1 K to up to 100 K. The analysis revealed that viewers showed little interest in informational and moralistic ads, but favored symbolic presentations of the Passion, the Resurrection, the Transfiguration, the Eucharist, the origin of evil, etc. The conclusion offers guidelines: the need to advertise, to adapt to audiences, to get feedback, and to preach through symbols rather than concepts. Full article

The paper aims to draft how phenomena such as abandonment, territorial disarticulation, environmental pollution, socioeconomic imbalances, and heritage consideration issues that surround landscapes where industrial activity has ceased are reflected on social media in Spain. The research focuses on the most popular social media platforms in Spain: Instagram, Facebook, and Twitter. A manual sample strategy was conducted to ensure an individualized approach to user-generated content. Sampling was carried out separately for three aspects: (1) keywords at a general level, (2) terms used to define industrial landscapes, and (3) recognition of significant industrial landscapes related to governmental facilities built in the 20th century, wherein we take into account three potential profile types: (i) individuals; (ii) NGOs/associations and/or public administrations; and (iii) academics. The results show that social media platforms are widely used as tools to disseminate information about industrial landscapes, but the contributions of each platform are uneven and incomplete in relation to the reality of post-industrial landscapes. However, it is worth recognizing the added value that their possible interaction brings as a reference for current civic debates. How social media contributes toward mitigating the difficulties of recognition, comprehension, and protection of post-industrial landscapes is emphasized in our conclusions. Full article

The current paper studies the 2022 parliamentary election campaign, in regards to what extent and quality certain elements of political debate can appear in political actors’ social media communication. During our research, we analyzed 2441 Facebook posts from parties and party leaders prior to the election. According to our results, political actors engage in opinionated discourse on social media and largely focus on public policy issues. They rarely rely on factual reasoning; instead, they tend to use individual phenomena to justify their claims. Ad hominem fallacy also plays a significant role in their Facebook posts when they are making an argument. However, other argumentation errors, so-called fallacies are quite rare in their communication. The main patterns are similar between the actors, but in general, parties and politicians from the opposition are more argumentative compared to the ruling party coalitions. Full article

Finances represent one of the key requirements to perform any useful activity for humanity. Financial markets, e.g., stock markets, forex, and mercantile exchanges, etc., provide the opportunity to anyone to invest and generate finances. However, to reap maximum benefits from these financial markets, effective decision making is required to identify the trade directions, e.g., going long/short by analyzing all the influential factors, e.g., price action, economic policies, and supply/demand estimation, in a timely manner. In this regard, analysis of the financial news and Twitter posts plays a significant role to predict the future behavior of financial markets, public sentiment estimation, and systematic/idiosyncratic risk estimation. In this paper, our proposed work aims to analyze the Twitter posts and Google Finance data to predict the future behavior of the stock markets (one of the key financial markets) in a particular time frame, i.e., hourly, daily, weekly, etc., through a novel StockSentiWordNet (SSWN) model. The proposed SSWN model extends the standard opinion lexicon named SentiWordNet (SWN) through the terms specifically related to the stock markets to train extreme learning machine (ELM) and recurrent neural network (RNN) for stock price prediction. The experiments are performed on two datasets, i.e., Sentiment140 and Twitter datasets, and achieved the accuracy value of 86.06%. Findings show that our work outperforms the state-of-the-art approaches with respect to overall accuracy. In future, we plan to enhance the capability of our method by adding other popular social media, e.g., Facebook and Google News etc. Full article

Alcohol advertising exposure is a risk factor for earlier alcohol initiation and higher alcohol consumption. Furthermore, engagement in digital alcohol marketing, such as liking or sharing an ad on social media, is associated with increased alcohol consumption and binge or hazardous drinking behavior. In light of these challenges, Lithuania has enacted a total prohibition on alcohol advertising, including social media. This study monitored the two most popular social media networks, Facebook and Instagram, to determine compliance with current legislation. In total, 64 Facebook and 51 Instagram profiles were examined. During the 60-day study period, 1442 and 749 posts on the selected Facebook and Instagram profiles, respectively, were published. There were a total of 163 distinct social media alcohol-related posts. Alcohol-related posts accounted for 5.9 percent of total Instagram posts and 8.3 percent of total Facebook posts. Alcohol advertisements accounted for 1.4 percent of all posts (infringement of the Alcohol Control Law). Influencers were responsible for nearly half (45.5 percent) of all observed alcohol-related Instagram posts. The study demonstrates high compliance with Lithuania’s total alcohol advertising ban on social media and emphasizes the importance of adequately monitoring the growing prominence of influencers on social media. Full article

Background: Undiagnosed human immunodeficiency virus (HIV) infection remains a public health challenge. We explore Facebook (FB) advertisement (Ads) cost per new HIV diagnosis using non-targeted Ads, a routine testing model against targeted Ads, and a focused testing model in Texas. Methods: On 14 October 2021, we created (without launching) Texas-based, USD 10 targeted (using criteria matching HIV populations at risk) and non-targeted FB Ads for 10 days. In the process of creating the Ads, we collected estimated audience size, daily reach, and daily clicks. We estimated Ad cost for each new HIV diagnosis for targeted and non-targeted Ads using new HIV diagnosis rates from focused and routine testing campaigns. Results: The Ad costs per new HIV diagnosis from the targeted model were 4.74, 2.86, 5.28, and 2.88 times lower for men, Black, Hispanic, and all age groups, respectively, when compared to the non-targeted model. The wider the gap was between new HIV diagnosis rates in a population for focused and routine testing, the more cost-effective targeted Ads became. Conclusions: Among HIV populations at risk, targeted FB Ads are more cost-effective for detecting new HIV infections than non-targeted Ads. This cost-effectiveness increases in locations where focused testing increases new HIV diagnosis rates, compared to routine testing. Full article