Search Results (74)

The Generalized Hough Transform (GHT) is a powerful method for rigid shape detection under rotation, scaling, translation, and partial occlusion conditions, but its four-dimensional accumulator incurs prohibitive computational and memory demands that prevent real-time deployment. To address this, we propose a framework that compresses the 4-D search space into a concise 2-D voting scheme by combining two-level sparse point-pair screening with an accelerated lookup. In the offline stage, template edges are extracted using an adaptive Canny operator with Otsu-determined thresholds, and gradient-direction differences for all point pairs are quantized to retain only those in the dominant bin, yielding rotation- and scale-invariant descriptors that populate a compact 2-D reference table. During the online stage, an adaptive grid selects only the highest-gradient pixels per cell as a base points, while a precomputed gradient-direction bucket table enables constant-time retrieval of compatible subpoints. Each valid base–subpoint pair is mapped to indices in the lookup table, and “fuzzy” votes are cast over a 3 × 3 neighborhood in the 2-D accumulator, whose global peak determines the object center. Evaluation on 200 real industrial parts—augmented to 1000 samples with noise, blur, occlusion, and nonlinear illumination—demonstrates that our method maintains over 90% localization accuracy, matches the classical GHT, and achieves a ten-fold speedup, outperforming IGHT and LI-GHT variants by 2–3×, thereby delivering a robust, real-time solution for industrial rigid object localization. Full article

Rape is typically committed as a one-on-one crime. However, a relatively high number of rapes (2–27%) involve a single victim and multiple perpetrators. These cases are often referred to as “gang” rapes but are also termed Multiple Perpetrator Rape (MPR). Despite these data, there is a scarce amount of legal decision-making research on this issue. This study investigated legal decision making in an acquaintance rape case involving multiple perpetrators. This study was a 2(Defendant Number: one vs. three) × 2(Victim Intoxication: intoxicated vs. sober) × 2(Participant Gender: women vs. men) between-participants design. Online community members (N = 171) were randomly assigned to read a trial summary involving one of four conditions. The primary results showed that, when the case involved multiple (vs. one) perpetrators, mock jurors were more likely to vote guilty, perceived the victim to be more helpless, and reported less sympathy for the defendant and lower defendant credibility. Cognitive networks showed that jurors in the MPR condition emphasized the number of perpetrators as a primary reason for voting guilty. Finally, there was evidence of a serial indirect effect involving victim helplessness and defendant blame that explained the relation between the number of defendants and verdicts, as well as parallel indirect effects of defendant credibility, sympathy, and anger, and victim helplessness on verdicts. Implications for prosecuting MPR cases are discussed. Full article

Threshold Multi-Party Private Set Intersection (TMP-PSI) is a cryptographic protocol that enables an element from the receiver’s set to be included in the intersection result if it appears in the sets of at least $t-1$ other participants, where t represents the threshold. This protocol is crucial for a variety of applications, such as anonymous electronic voting, online ride-sharing, and close-contact tracing programs. However, most existing TMP-PSI schemes are designed based on threshold homomorphic encryption, which faces significant challenges, including low computational efficiency and a high number of communication rounds. To overcome these limitations, this study introduces the Threshold Oblivious Pseudo-Random Function (tOPRF) to fulfill the requirements of threshold encryption and decryption. Additionally, we extend the concept of the Oblivious Programmable Pseudo-Random Function (OPPRF) to develop a novel cryptographic primitive termed the Partially OPPRF (P-OPPRF). This new primitive retains the critical properties of obliviousness and randomness, along with the security assurances inherited from the OPPRF, while also offering strong resistance against malicious adversaries. Leveraging this primitive, we propose the first malicious-secure TMP-PSI protocol, named QMP-PSI, specifically designed for applications like anonymous electronic voting systems. The protocol effectively counters collusion attacks among multiple parties, ensuring robust security in multi-party environments. To further enhance voting efficiency, this work presents a cloud-assisted QMP-PSI to outsource the computationally intensive phases. This ensures that the computational overhead for participants is solely dependent on the set size and statistical security parameters, thereby maintaining security while significantly reducing the computational burden on voting participants. Finally, this work validates the protocol’s performance through extensive experiments under various set sizes, participant numbers, and threshold values. The results demonstrate that the protocol surpasses existing schemes, achieving state-of-the-art (SOTA) performance in communication overhead. Notably, in small-scale voting scenarios, it exhibits exceptional performance, particularly when the threshold is small or close to the number of participants. Full article

In 2023, Brazil regained momentum in proposing innovative participatory institutions by launching a complex participatory experiment for budget planning within its “Multiannual Plan” (PPA). While this was not a scaled-up version of the local participatory budgeting plan that emerged decades earlier, its launch did reopen the debate on the possibilities of expanding political participation. The challenge was significant due to the intricate nature of budget planning and the complexity of the plan’s participatory design. This paper examines that experience by outlining the institutional design of the PPA Participativo and analyzing its results in terms of online participation. It also discusses the prominence of climate-related proposals, suggesting that the PPA Participativo serves as a relevant indicator of national concerns regarding climate-based issues. Building on previous experiences, the PPA Participativo is a strategy consisting of three layers: an online platform for digital participation, state-level meetings with civil society activists, and a high-level forum composed of members from national councils. This paper analyses some of the results from the online platform, which recorded 4 million visits from 1.5 million individual participants. These citizens were invited to submit proposals, vote for up to three proposals, and choose from a set of pre-designed government programs. The main concerns that emerged from this open-ended process included corporate demands, stemming from highly organized sectors, such as public health and education employees. However, animal rights also ranked among the most-voted proposals. The program of the Ministry of the Environment on climate change reached the top position in this segment, largely due to its strong campaigning strategy. This paper discusses these outcomes, drawing on evidence of both societal engagement and institutional activism to promote specific agendas. Full article

In the digital era, social media platforms have seen a substantial increase in the volume of online comments. While these platforms provide users with a space to express their opinions, they also serve as fertile ground for the proliferation of hate speech. Hate comments can be categorized into various types, including discrimination, violence, racism, and sexism, all of which can negatively impact mental health. Among these, sexism poses a significant challenge due to its various forms and the difficulty in defining it, making detection complex. Nevertheless, detecting and preventing sexism on social networks remains a critical issue. Recent studies have leveraged language models such as transformers, known for their ability to capture the semantic nuances of textual data. In this study, we explore different transformer models, including multiple versions of RoBERTa (A Robustly Optimized BERT Pretraining Approach), to detect sexism. We hypothesize that combining a sentiment-focused language model with models specialized in sexism detection can improve overall performance. To test this hypothesis, we developed two approaches. The first involved using classical transformers trained on our dataset, while the second combined embeddings generated by transformers with a Long Short-Term Memory (LSTM) model for classification. The probabilistic outputs of each approach were aggregated through various voting strategies to enhance detection accuracy. The LSTM with embeddings approach improved the F1-score by 0.2% compared to the classical transformer approach. Furthermore, the combination of both approaches confirms our hypothesis, achieving a 1.6% improvement in the F1-score in each case. We determined that an F1 score of over 0.84 effectively measures sexism. Additionally, we constructed our own dataset to train and evaluate the models. Full article

In online social networks, users can vote on different trust levels for each other to indicate how much they trust their friends. Researchers have improved their ability to predict social trust relationships through a variety of methods, one of which is the graph neural network (GNN) method, but they have also brought the vulnerability of the GNN method into the social trust network model. We propose a data-poisoning attack method for GNN-based social trust models based on the characteristics of social trust networks. We used a two-sample test for power-law distributions of discrete data to avoid changes in the dataset being detected and used an enhanced surrogate model to generate poisoned samples. We further tested the effectiveness of our approach on three real-world datasets and compared it with two other methods. The experimental results using three datasets show that our method can effectively avoid detection. We also used three metrics to illustrate the effectiveness of our attack, and the experimental results show that our attack stayed ahead of the other two methods in all three datasets. In terms of one of our metrics, our attack method decreased the accuracies of the attacked models by 12.6%, 22.8%, and 13.8%. Full article

Cyberbullying is a serious problem in online communication. It is important to find effective ways to detect cyberbullying content to make online environments safer. In this paper, we investigated the identification of cyberbullying contents from the Bangla and Chittagonian languages, which are both low-resource languages, with the latter being an extremely low-resource language. In the study, we used both traditional baseline machine learning methods, as well as a wide suite of deep learning methods especially focusing on hybrid networks and transformer-based multilingual models. For the data, we collected over 5000 both Bangla and Chittagonian text samples from social media. Krippendorff’s alpha and Cohen’s kappa were used to measure the reliability of the dataset annotations. Traditional machine learning methods used in this research achieved accuracies ranging from 0.63 to 0.711, with SVM emerging as the top performer. Furthermore, employing ensemble models such as Bagging with 0.70 accuracy, Boosting with 0.69 accuracy, and Voting with 0.72 accuracy yielded promising results. In contrast, deep learning models, notably CNN, achieved accuracies ranging from 0.69 to 0.811, thus outperforming traditional ML approaches, with CNN exhibiting the highest accuracy. We also proposed a series of hybrid network-based models, including BiLSTM+GRU with an accuracy of 0.799, CNN+LSTM with 0.801 accuracy, CNN+BiLSTM with 0.78 accuracy, and CNN+GRU with 0.804 accuracy. Notably, the most complex model, (CNN+LSTM)+BiLSTM, attained an accuracy of 0.82, thus showcasing the efficacy of hybrid architectures. Furthermore, we explored transformer-based models, such as XLM-Roberta with 0.841 accuracy, Bangla BERT with 0.822 accuracy, Multilingual BERT with 0.821 accuracy, BERT with 0.82 accuracy, and Bangla ELECTRA with 0.785 accuracy, which showed significantly enhanced accuracy levels. Our analysis demonstrates that deep learning methods can be highly effective in addressing the pervasive issue of cyberbullying in several different linguistic contexts. We show that transformer models can efficiently circumvent the language dependence problem that plagues conventional transfer learning methods. Our findings suggest that hybrid approaches and transformer-based embeddings can effectively tackle the problem of cyberbullying across online platforms. Full article

The SUNshiners group includes people in the early stages of dementia with an interest in dementia activism and research. The group found that despite the growing awareness of invisible disabilities, there is very limited research into the pros and cons of the invisibility of dementia. Our paper explores the SUNshiners research which stemmed from varied individual experiences of disclosing diagnoses. The group designed and developed a short survey to explore what the public knew about dementia and what they thought about the invisibility of dementia. A mixture of open- and closed-ended questions were used to gain meaningful data. A total of 347 people completed the survey (315 online and 32 paper-based), which was then co-analysed. The findings suggest that the majority of the public felt that the invisibility of dementia was negative; that knowing someone had dementia when first meeting them would be beneficial; that people living with dementia should maintain the right to vote; and that people living with dementia do not automatically require a consistent, regular carer. Common themes from the open-ended answers included capacity, severity of dementia, and access to support. The findings support the disclosure of dementia diagnosis; however, more action is needed to tackle stigmatised views, particularly as the SUNshiners felt that people do not have enough dementia education to support a positive disclosure experience. They shared their experiences of the group and the project’s benefits, but also the losses they have faced. Our paper aims to be as accessible as possible. Full article

Document image retrieval (DIR) systems simplify access to digital data within printed documents by capturing images. These systems act as bridges between print and digital realms, with demand in organizations handling both formats. In education, students use DIR to access online materials, clarify topics, and find solutions in printed textbooks by photographing content with their phones. DIR excels in handling complex figures and formulas. We propose using ensembles of DIR systems instead of single-feature models to enhance DIR’s efficacy. We introduce “Vote-Based DIR” and “The Strong Decision-Based DIR”. These ensembles combine various techniques, like optical code reading, spatial analysis, and image features, improving document retrieval. Our study, using a dataset of university exam preparation materials, shows that ensemble DIR systems outperform individual ones, promising better accuracy and efficiency in digitizing printed content, which is especially beneficial in education. Full article

Sub-Saharan Africa is facing an expected doubling of human population and tripling of food demand over the next quarter century, posing a range of severe environmental, political, and socio-economic challenges. In some cases, key Sustainable Development Goals (SDGs) are in direct conflict, raising difficult policy and funding decisions, particularly in relation to trade-offs between food production, social inequality, and ecosystem health. In this study, we used a horizon-scanning approach to identify 100 practical or research-focused questions that, if answered, would have the greatest positive impact on addressing these trade-offs and ensuring future productivity and resilience of food-production systems across sub-Saharan Africa. Through direct canvassing of opinions, we obtained 1339 questions from 331 experts based in 55 countries. We then used online voting and participatory workshops to produce a final list of 100 questions divided into 12 thematic sections spanning topics from gender inequality to technological adoption and climate change. Using data on the background of respondents, we show that perspectives and priorities can vary, but they are largely consistent across different professional and geographical contexts. We hope these questions provide a template for establishing new research directions and prioritising funding decisions in sub-Saharan Africa. Full article

In recent years, advances in deep learning (DL) techniques for video analysis have developed to solve the problem of real-time processing. Automated face recognition in the runtime environment has become necessary in video surveillance systems for urban security. This is a difficult task due to face occlusion, which makes it hard to capture effective features. Existing work focuses on improving performance while ignoring issues like a small dataset, high computational complexity, and a lack of lightweight and efficient feature descriptors. In this paper, face recognition (FR) using a Convolutional mixer (AFR-Conv) algorithm is developed to handle face occlusion problems. A novel AFR-Conv architecture is designed by assigning priority-based weight to the different face patches along with residual connections and an AdaBoost classifier for automatically recognizing human faces. The AFR-Conv also leverages the strengths of pre-trained CNNs by extracting features using ResNet-50, Inception-v3, and DenseNet-161. The AdaBoost classifier combines these features’ weighted votes to predict labels for testing images. To develop this system, we use the data augmentation method to enhance the number of datasets using human face images. The AFR-Conv method is then used to extract robust features from images. Finally, to recognize human identity, an AdaBoost classifier is utilized. For the training and evaluation of the AFR-Conv model, a set of face images is collected from online data sources. The experimental results of the AFR-Conv approach are presented in terms of precision (PR), recall (RE), detection accuracy (DA), and F1-score metrics. Particularly, the proposed approach attains 95.5% PR, 97.6% RE, 97.5% DA, and 98.5% of F1-score on 8500 face images. The experimental results show that our proposed scheme outperforms advanced methods for face classification. Full article

With the advent of the digital age, traditional lifestyle activities, such as reading books, referencing recipes, and enjoying music, have progressively transitioned from offline to online. However, numerous issues plague the conventional approach to digital copyright protection. This is especially true in the realm of recipe protection, where the rights and interests of original creators are inadequately safeguarded due to the widespread dissemination of a large number of recipes on the Internet. This primarily stems from the high costs of gathering evidence, incomplete coverage of evidence collection, and the inability to identify and halt infringement activities in a timely manner during the process of traditional digital copyright protection. Therefore, this study designs and implements a blockchain-based digital recipe copyright protection scheme to address the issues of insufficient legal evidence and cumbersome processes in traditional digital copyright protection. First, we enhance standard short text similarity calculation method SimHash, boosting the accuracy of text similarity detection. We then utilize the decentralization, immutability, time-stamping, traceability, and smart contract features of blockchain technology for data privacy protection. We employ the Interplanetary File System (IPFS) to store raw data, thereby ensuring user privacy and security. Lastly, we improve the proxy voting node selection in the existing delegated proof of stake (DPOS) consensus mechanism. According thorough evaluation and empirical analysis, the scheme effectively improves the accuracy of text similarity detection. Simultaneously, the enhanced DPOS mechanism effectively rewards nodes with excellent performance and penalizes nodes exhibiting malicious behavior. In this study, we successfully designed and implemented an innovative digital recipe copyright protection scheme. This scheme effectively enhances the accuracy of text similarity detection; ensures the privacy and security of user data; and, through an enhanced DPOS mechanism, rewards well-performing nodes while penalizing those exhibiting malicious behavior. Full article

Since online social networks play an increasingly important role in the final voting decision of each individual, political parties and candidates are changing the way of doing politics and campaigning, increasing their digital presence. In this paper, we propose a methodology to analyze and measure the emotions that news agencies express on social media towards candidates and apply it to the 2018 Brazilian elections. The presented method is based on a sentiment analysis and emotion mining by means of machine learning and Natural Language Processing approaches such as Naïve Bayes classification and Stemming calculation. We found that if doing basic sentiment detection, nearly all posts are neutral. However, when we analyze emotions, following Ekman’s six basic emotions, we do not find neutrality but clear and identifiable emotions. Next, we present and discuss the associative patterns between news agencies and presidential candidates. Finally, since the candidate that captured the highest and most negative attention emerged victorious in the elections, we discuss the potential importance of having a social media presence, regardless of generating positive or negative emotions. Full article

Introduction: There are currently three first-line immunotherapy options used as monotherapy in advanced non-small cell lung cancer (NSCLC) patients with high programmed death ligand 1 (PD-L1) expression (≥50%). This manuscript aims to evaluate the available data on atezolizumab (AT), cemiplimab (CEMI), and pembrolizumab (PEMBRO) and to study the results obtained during pivotal trials, especially regarding patient subgroups. Methods: Nominal group and Delphi techniques were used. Eight Spanish experts in lung cancer (the scientific committee of the project) analyzed the use of immunotherapy monotherapy as first-line treatment in patients with NSCLC and high PD-L1 expression. The expert scientific committee formulated several statements based on a scientific review and their own clinical experience. Subsequently, 17 additional Spanish lung cancer experts were selected to appraise the committee’s statements through two Delphi rounds. They completed a Delphi round via an online platform and voted according to a scale from 1 (strongly disagree) to 10 (strongly agree). The statements were approved if ≥70% of experts voted 7 or more. Results: A total of 20 statements were proposed covering the following areas: (1) general characteristics of pivotal clinical trials; (2) overall main outcomes of pivotal clinical trials; and (3) subgroup analysis. All statements reached consensus in the first round. Conclusions: AT, CEMI, and PEMBRO as monotherapy can be considered the standard of care in patients with advanced NSCLC and high PD-L1 expression (≥50%). Moreover, some differences noted among the drugs analyzed in this document might facilitate treatment decision-making, especially in clinically relevant patient subgroups, when using PD-1/PD-L1 inhibitors. The high level of agreement reached among experts supports the proposed statements. Full article

The current traditional paper ballot voting schemes suffer from several limitations such as processing delays due to counting paper ballots, lack of transparency, and manipulation of the ballots. To solve these limitations, an electronic voting (e-voting) scheme has received massive interest from both governments and academia. In e-voting, individuals can cast their vote online using their smartphones without the need to wait in long lines. Additionally, handicapped voters who face limited wheelchair access in many polling centers could now participate in elections hassle-free. The existing e-voting schemes suffer from several limitations as they are either centralized, based on public blockchains, or utilize local private blockchains. This results in privacy issues (using public blockchains) or large financial costs (using local/private blockchains) due to the amount of computing power and technical knowledge needed to host blockchains locally. To address the aforementioned limitations, in this paper, we propose an online voting scheme using IBM cloud-based Hyperledger Fabric. Our scheme allows voters to cast their encrypted votes in a secure manner. Then any participant can obtain the ballot results in a decentralized and transparent manner, without sacrificing the privacy of individual voters. We implement the proposed scheme using IBM cloud-based Hyperledger Fabric. The experimental results identify the performance characteristics of our scheme and demonstrate that it is feasible to run an election consisting of thousands of participants using cloud-based Fabric. Full article