Recommender Systems (RS) influence everyday decisions, yet most remain optimized for short-term engagement or commercial gain. RS4SD aims to shift this focus by exploring how RS can contribute to sustainable development through behavioral change and nudging strategies. Aligned with the UN Sustainable Development Goals (SDG), RS4SD will highlight applications that promote responsible consumption, sustainable mobility, healthy eating, and digital well-being. In particular, we will focus on how AI and RS can be designed to foster sustainable behaviors through multi-objective optimization and ethically aligned interventions. These objectives are directly tied to the UN SDG, and we welcome all contributions showcasing RS in support of these goals. A central theme of the workshop is the integration of behavioral science and AI to design interventions that guide users toward more sustainable and healthier choices while preserving individual autonomy. Topics of interest include multi-objective recommendation, health-aware RS, eco-friendly product and tourism RS, as well as novel evaluation metrics that go beyond accuracy to capture societal impact. RS4SD will bring together researchers, stakeholders and practitioners from RS, AI, sustainability, and behavioral science to share models, datasets, frameworks, and real-world use cases. The workshop encourages interdisciplinary collaboration and aims to build a community dedicated to responsible, behavior-aware RS that benefit both individuals and society.

Abstract Tuberculosis (TB) is a highly contagious disease that affects millions of individuals globally. Early detection is crucial in preventing its spread and improving patient outcomes. Radiologists often utilize X-ray imaging as a diagnostic tool for tuberculosis; however, the accuracy of the results may differ depending on the radiologist’s interpretation. To increase the precision of TB identification from X-ray images, a hybrid strategy combining Convolutional Neural Network, Histogram of Oriented Gradients, and Quantum Support Vector Machine (QSVM) classifier has been developed. Quantum machine learning is a rapidly growing field at the intersection of quantum computing and machine learning. By combining the strengths of both techniques, this approach aims to capture relevant features and shape and texture information of an image. The study used state-of-the-art models such as VGG16 and AlexNet, along with a handcrafted method (HOG) and a deep learning (CNN) method to detect TB from chest X-ray images. Six distinct tests were carried out by the authors to successfully diagnose tuberculosis. All models showed promise when the data was analyzed, but the best techniques involved augmentation, image preprocessing, contrast improvement, and noise filtering. The block-matching and 3D filtering approach was used to improve image edge preservation and reduce noise. In this study, Grad-CAM (gradient-weighted class activation map) was applied to the convolutional neural network model to identify the model’s important features (explainability) for decision-making. This focus on Explainable AI (XAI) is crucial for clinical adoption, as it provides radiologists with visual evidence of the model’s reasoning, fostering trust and enabling more informed decisions. The hybrid model was found to have achieved an exceptional level of accuracy in detecting and classifying TB, with an accuracy rate of 100% during training, 99.07% during validation, and 98.14% during testing. These results highlight the effectiveness of the hybrid model in accurately identifying TB and its potential to be a valuable tool in the fight against this disease.

Abstract Airborne infectious diseases are a significant threat to human beings. Nowadays, one of the deadliest airborne diseases, coronavirus (COVID-19), is resulting in a massive health crisis due to its rapid transmission. The World Health Organization for protection against the spread of airborne diseases has set several guidelines. The most effective preventive measure against airborne diseases, according to the World Health Organization, is wearing masks in public places and crowded areas. It is challenging to monitor people manually in these areas. In this study, we collect data from public and local sources to develop an occlusion-aware face mask detection model. This study presents a deep learning-based occlusion-aware face mask detection model designed to identify both proper and improper mask usage, even under partial facial occlusions. A dataset of 4,820 images, including occlusions from hands, objects, and mask misuse, was used to train and evaluate three convolutional neural network models: InceptionV3, MobileNetV2, and DenseNet121. Among them, DenseNet121 achieved the highest accuracy of 96.3% on test data. Therefore, our proposed study is used to investigate occlusion aware face mask classification using deep learning.

Abstract Food recommendation systems have become increasingly popular due to the proliferation of online food service websites. Accordingly, the ratings assigned by users are one of the most important resources in these systems. However, users generally express their opinions about a few foods, which results in data sparsity. Furthermore, food recommendation is a health-critical task, as recommending unhealthy foods to users may threaten their health. In this paper, we developed a novel rating profile expansion approach for food recommenders that considers both health and reliability measures. This approach enhances the efficiency of the user’s rating profile by including healthy and reliable virtual ratings. Specifically, we introduce a probabilistic rating profile evaluation technique to determine whether a profile needs to be expanded. Then, those profiles with an insufficient number of ratings are automatically expanded by adding virtual ratings obtained using the opinions of users who belong to the target user’s community. For this purpose, the users are grouped using a novel time-aware community detection algorithm based on their preferences. Moreover, a health-aware reliability measure is proposed so that only the most reliable virtual ratings are accounted for in the target user’s rating profile expansion. Therefore, the developed approach not only mitigates issues stemming from sparse data in food recommendation systems but also makes them more effective in recommending healthy foods to users. Experiments conducted on two publicly available real-world datasets demonstrated that the developed system is superior to other baseline models.

Standard evaluation of automated text summarization (ATS) methods relies on manually crafted golden summaries. With the advances in Large Language Models (LLMs), it is legitimate to question whether these models can now potentially complement or replace human-crafted summaries. This study examines the effectiveness of several language models (LMs) in specifically addressing the issue of preserving factual consistency. By conducting a thorough assessment of various conventional and state-of-the-art performance metrics, such as ROUGE, BLEU, BERTScore, FActScore, and LongDocFACTScore across diverse datasets, our findings highlight the important relationship between linguistic eloquence and factual accuracy. The findings suggest that whereas LLMs, such as GPT and LLaMA, demonstrate considerable competence in producing concise and contextually-aware summaries, there remain difficulties in ensuring factual accuracy, particularly in domain-specific situations. Moreover, this work enhances the existing knowledge on summarization dynamics and highlights the need of developing more reliable and tailored evaluation techniques that minimize the probability of factual errors in text generated by ATS. In particular, the findings advance the current domain by providing a rigorous assessment of the balance between linguistic fluency and factual correct- ness, highlighting the limitations of current ATS frameworks and metrics to enhance the factual reliability of LM-generated summaries.

The use of video surveillance in public spaces has increased rapidly, extending to residential areas and even nonmetropolitan regions. However, there is limited research on public attitudes toward CCTV, and few studies have focused on how these attitudes have changed over the years. This study examines public sentiment toward surveillance cameras from 2000 to 2020 using data from Suomi24, Finland’s largest forum. We begin by formulating a set of pertinent hypotheses and then employ the advanced Snowflake-artic-embedding model to validate these hypotheses. Simultaneously, exploratory data analysis is conducted on the collected dataset to understand trends, thematic distribution, the impact of the GDPR introduction, the effects of Snowden’s revelations, and the overall sentiment polarity. The analysis revealed key shifts in sentiment, particularly during global events like the Snowden revelations and the introduction of the GDPR. Moreover, the findings highlight a complex balance between privacy, security, and surveillance, providing insights for policymakers and technology developers aiming to navigate public concerns around surveillance practices.

This paper reports on the DISH4U, a mobile app designed to guide the users towards healthy foods with a special focus on Finnish market. The application makes use of the Finnish Institute of Health and Welfare (THL) nutritional database Fineli to determine the nutritional content of the food. Besides, the app takes into account user’s profile in terms of food preferences, health and sport activities, and then elucidating the user about nutritional content of food served at selected restaurants as well as providing healthy recommendations available at nearby restaurants.

This paper explores the development of a software code explanation generation tool using CodeLlema 13B Instruct model under zero-shot learning scheme. The generated code explanations are designed to enhance problem localization and comprehension within Nokia Mobile Networks Solutions’ SoC codebases. We assess the semantics and readability of the generated explanations through a human-annotated dataset crafted by SoC engineers within the Nokia group alongside publicly available Code-NL pair dataset from the CoNala Corpus.

Food recommendation systems play a crucial role in promoting personalized recommendations designed to help users find food and recipes that align with their preferences. However, many existing food recommendation systems have overlooked the important aspect of healthy-food and nutritional value of recommended foods, thereby limiting their effectiveness in generating truly healthy recommendations. Our preliminary analysis indicates that users tend to respond positively to unhealthy food and recipes. As a result, existing food recommender systems that neglect health considerations often assign high scores to popular items, inadvertently encouraging unhealthy choices among users. In this study, we propose the development of a fairness-based model that prioritizes health considerations. Our model incorporates fairness constraints from both the user and item perspectives, integrating them into a joint objective framework. Experimental results conducted on real-world food datasets demonstrate that the proposed system not only maintains the ability of food recommendation systems to suggest users’ favorite foods but also improves the health factor compared to unfair models, with an average enhancement of approximately 35%.

Hybrid recommender systems utilize advanced algorithms capable of learning heterogeneous sources of data and generating personalized recommendations for users. The data can range from user preferences (e.g., ratings or reviews) to item content (e.g., description or category). Prior studies in the field of recommender systems have primarily relied on ”ratings” as the user feedback, when building user profiles or evaluating the quality of the recommendation. While ratings are informative, they may still fail to represent a comprehensive picture of actual user preferences. In contrast, there are other types of feedback data that differently or complementarily represent users and their preferences, including the reviews and the sentiments encapsulated within them. Such data can reveal important parts of a user’s profile that are not necessarily correlated with user ratings, and hence, they potentially reflect a different side of the user’s profile. In this paper, we propose a novel form of hybrid recommender system, capable of analyzing the reviews and extracting their sentiments that are incorporated into the recommendation process. We used advanced algorithms to generate recommendations for users capable of incorporating additional data, such as the review sentiment. We conducted analyses and showed that sentiments of user reviews are not always highly correlated with the ratings (e.g., in music domain). This might mean that sentiment can be indicative of a different aspect of user preferences and can be used as an alternative signal of user feedback. Hence, we have used both ratings and sentiments of reviews when evaluating our proposed hybrid recommender system. We selected two common datasets for the evaluation, Amazon Digital Music and Amazon Video Games, and showed the superior performance of the proposed hybrid recommender system compared to different baselines. The comparison were made in two evaluation scenarios, namely, when the ratings were considered the user feedback and when sentiments of the review were considered the user feedback.

Abstract Cross-border trade barriers introduced by national authorities to protect local business and labor force cause substantial damage to international economical actors. Therefore, identifying such barriers beyond regulator’s audit reporting is of paramount importance. This paper contributes towards this goal by proposing a novel approach that uses natural language processing and deep learning method for uncovering Finnish-Russian trade barriers in the fish industry from selected business discussion forums. Especially, the approach makes use i) a three-leg ontology for data collection, ii) a BERT architecture for mapping Onkivisit-Shaw-Kananen trade barrier ontology to negative polarity posts and, iii) a new reverse-engineering clustering approach to identify the causes of individual trade-barrier types. A comparison with official statistical reports has been carried out to identify the salient aspects of trade-barriers that hold regardless of the time difference. The findings reveal the dominance of the Time-length barrier type in the Finnish discussion forum dataset and import vs export tariff discrepancy and product requirement barrier types in the Russian forum dataset. The developed framework can serve as a tool to assist companies or regulators in providing business-related recommendations to overcome the detected trade barriers.

COVID-19 pandemic has fueled the interest in artificial intelligence tools for quick diagnosis to limit virus spreading. Over 60% of people who are infected complain of a dry cough. Cough and other respiratory sounds were used to build diagnosis models in much recent research. We propose in this work, an augmentation pipeline which is applied on the pre-filtered data and uses i) pitch-shifting technique to augment the raw signal and, ii) spectral data augmentation technique SpecAugment to augment the computed mel-spectrograms. A deep learning based architecture that hybridizes convolution neural networks and long-short term memory with an attention mechanism is proposed for building the classification model. The feasibility of the proposed is demonstrated through a set of testing scenarios using the large-scale COUGHVID cough dataset and through a comparison with three baselines models. We have shown that our classification model achieved 91.13% of testing accuracy, 90.93% of sensitivity and an area under the curve of receiver operating characteristic of 91.13%.

In the era of social media and mobile internet, the design of automatic tools for online detection of hate speech and/or abusive language becomes crucial for society and community empowerment. Nowadays of current technology in this respect is still limited and many service providers are still relying on the manual check. This paper aims to advance in this topic by leveraging novel natural language processing, machine learning, and feature engineering techniques. The proposed approach advocates a classification-like technique that makes use of a special data design procedure. The latter enforces a balanced training scheme by exploring the negativity of the original dataset. This generates new transfer learning paradigms, Two classification schemes using convolution neural network and LSTN architecture that use FastText embeddings as input features are contrasted with baseline models constituted of Logistic regression and Naives' Bayes classifiers. Wikipedia Comment dataset constituted of Personal Attack, Aggression and Toxicity data are employed to test the validity and usefulness of the proposal.

When a terror-related event occurs, there is a surge of traffic on social media comprising of informative messages, emotional outbursts, helpful safety tips, and rumors. It is important to understand the behavior manifested on social media sites to gain a better understanding of how to govern and manage in a time of crisis. We undertook a detailed study of Twitter during two recent terror-related events: the Manchester attacks and the Las Vegas shooting. We analyze the tweets during these periods using (a) sentiment analysis, (b) topic analysis, and (c) fake news detection. Our analysis demonstrates the spectrum of emotions evinced in reaction and the way those reactions spread over the event timeline. Also, with respect to topic analysis, we find “echo chambers”, groups of people interested in similar aspects of the event. Encouraged by our results on these two event datasets, the paper seeks to enable a holistic analysis of social media messages in a time of crisis.

How can we identify individuals at risk of being drawn into online sex work? The spread of online communication removes transaction costs and enables a greater number of people to be involved in illicit activities, including online sex trade. As a result, social media platforms often work as springboard for criminal careers posing a significant risk to the economy, public health and trust. Detecting deviant behaviors online is limited by the poor availability of ground-truth data and machine learning tools. Unlike prior work which focuses exclusively on either qualitative or quantitative methods, in this paper we combine covert online ethnography with semi-supervised learning methodologies, using data from a popular European adult forum. We obtained risk assessment results of 78 users using covert online ethnography, and set out to build a machine learning model that can predict the risk factor in other 28,832 users. Results show that a combination-based approach in which all features are used yields the most accurate results.

This paper proposes an effective way to discover and memorize new English vocabulary based on both semantic and phonetic associations. The method we proposed aims to automatically find out the most associated words of a given target word. The measurement of semantic association was achieved by calculating cosine similarity of two-word vectors, and the measurement of phonetic association was achieved by calculating the longest common subsequence of phonetic symbol strings of two words. Finally, the method was implemented as a web application.

This article attempts to analyze the combination of uncertain pieces of information, particularly, given several pieces of information, each of which is assigned a certainty factor. The problem then is whether it is possible to find the certainty factor associated with the combination result. This problem, even if it has been widely commented on in probability literature from various viewpoints, sounds less analyzed in the framework of possibility or evidence theories. This study investigates various proposals for quantifying certainty qualification and then constructs the certainty assigned for the combination result of initial inputs. First, we shall consider some basic combination modes and then attempt to generalize the result for more general combination modes. © 2004 Wiley Periodicals, Inc.