arXiv subjects

Noel Crespi

Publications and source records attributed to Noel Crespi.

At least 37 records · Page 2Linked to original sources

arXiv2023

On the definition of toxicity in NLP

The fundamental problem in toxicity detection task lies in the fact that the toxicity is ill-defined. This causes us to rely on subjective and vague data in models' training, which results in non-robust and non-accurate results: garbage in - garbage out. This work suggests a new, stress-level-based definition of toxicity designed to be objective and context-aware. On par with it, we also describe possible ways of applying this new definition to dataset creation and model training.

cs.CL↗

arXiv2023

No offence, Bert -- I insult only humans! Multiple addressees sentence-level attack on toxicity detection neural network

We introduce a simple yet efficient sentence-level attack on black-box toxicity detector models. By adding several positive words or sentences to the end of a hateful message, we are able to change the prediction of a neural network and pass the toxicity detection system check. This approach is shown to be working on seven languages from three different language families. We also describe the defence mechanism against the aforementioned attack and discuss its limitations.

cs.CL↗

arXiv2023

Hate Speech and Offensive Language Detection using an Emotion-aware Shared Encoder

The rise of emergence of social media platforms has fundamentally altered how people communicate, and among the results of these developments is an increase in online use of abusive content. Therefore, automatically detecting this content is essential for banning inappropriate information, and reducing toxicity and violence on social media platforms. The existing works on hate speech and offensive language detection produce promising results based on pre-trained transformer models, however, they considered only the analysis of abusive content features generated through annotated datasets. This paper addresses a multi-task joint learning approach which combines external emotional features extracted from another corpora in dealing with the imbalanced and scarcity of labeled datasets. Our analysis are using two well-known Transformer-based models, BERT and mBERT, where the later is used to address abusive content detection in multi-lingual scenarios. Our model jointly learns abusive content detection with emotional features by sharing representations through transformers' shared encoder. This approach increases data efficiency, reduce overfitting via shared representations, and ensure fast learning by leveraging auxiliary information. Our findings demonstrate that emotional knowledge helps to more reliably identify hate speech and offensive language across datasets. Our hate speech detection Multi-task model exhibited 3% performance improvement over baseline models, but the performance of multi-task models were not significant for offensive language detection task. More interestingly, in both tasks, multi-task models exhibits less false positive errors compared to single task scenario.

cs.CL↗

arXiv2022

BERT-based Ensemble Approaches for Hate Speech Detection

With the freedom of communication provided in online social media, hate speech has increasingly generated. This leads to cyber conflicts affecting social life at the individual and national levels. As a result, hateful content classification is becoming increasingly demanded for filtering hate content before being sent to the social networks. This paper focuses on classifying hate speech in social media using multiple deep models that are implemented by integrating recent transformer-based language models such as BERT, and neural networks. To improve the classification performances, we evaluated with several ensemble techniques, including soft voting, maximum value, hard voting and stacking. We used three publicly available Twitter datasets (Davidson, HatEval2019, OLID) that are generated to identify offensive languages. We fused all these datasets to generate a single dataset (DHO dataset), which is more balanced across different labels, to perform multi-label classification. Our experiments have been held on Davidson dataset and the DHO corpora. The later gave the best overall results, especially F1 macro score, even it required more resources (time execution and memory). The experiments have shown good results especially the ensemble models, where stacking gave F1 score of 97% on Davidson dataset and aggregating ensembles 77% on the DHO dataset.

cs.CL↗

arXiv2021

Fabric-SCF: A Blockchain-based Secure Storage and Access Control Scheme for Supply Chain Finance

Supply chain finance(SCF) is committed to providing credit for small and medium-sized enterprises(SMEs) with low credit lines and small financing scales. The resulting financial credit data and related business transaction data are highly confidential and private. However, traditional SCF management schemes mostly use third-party platforms and centralized designs, which cannot achieve highly reliable secure storage and fine-grained access control. To fill this gap, this paper designs and implements Fabric-SCF, a secure storage and access control system based on blockchain and attribute-based access control (\textbf{ABAC}) model. This scheme uses distributed consensus to realize data security, traceability, and immutability. We also use smart contracts to define system processes and access policies to ensure the efficient operation of the system. To verify the performance of Fabric-SCF, we designed two sets of simulation experiments. The results show that Fabric-SCF achieves dynamic and fine-grained access control while maintaining high throughput in a simulated real-world operating scenario.

cs.CR↗

arXiv2020

Characterising and Detecting Sponsored Influencer Posts on Instagram

Recent years have seen a new form of advertisement campaigns emerge: those involving so-called social media influencers. These influencers accept money in return for promoting products via their social media feeds. Although this constitutes a new and interesting form of marketing, it also raises many questions, particularly related to transparency and regulation. For example, it can sometimes be unclear which accounts are officially influencers, or what even constitutes an influencer/advert. This is important in order to establish the integrity of influencers and to ensure compliance with advertisement regulation. We gather a large-scale Instagram dataset covering thousands of accounts advertising products, and create a categorisation based on the number of users they reach. We then provide a detailed analysis of the types of products being advertised by these accounts, their potential reach, and the engagement they receive from their followers. Based on our findings, we train machine learning models to distinguish sponsored content from non-sponsored, and identify cases where people are generating sponsored posts without officially labelling them. Our findings provide a first step towards understanding the under-studied space of online influencers that could be useful for researchers, marketers and policymakers.

cs.SI↗

arXiv2020

Impersonation on Social Media: A Deep Neural Approach to Identify Ingenuine Content

Impersonators are playing an important role in the production and propagation of the content on Online Social Networks, notably on Instagram. These entities are nefarious fake accounts that intend to disguise a legitimate account by making similar profiles and then striking social media by fake content, which makes it considerably harder to understand which posts are genuinely produced. In this study, we focus on three important communities with legitimate verified accounts. Among them, we identify a collection of 2.2K impersonator profiles with nearly 10k generated posts, 68K comments, and 90K likes. Then, based on profile characteristics and user behaviours, we cluster them into two collections of `bot' and `fan'. In order to separate the impersonator-generated post from genuine content, we propose a Deep Neural Network architecture that measures `profiles' and `posts' features to predict the content type: `bot-generated', 'fan-generated', or `genuine' content. Our study shed light into this interesting phenomena and provides interesting observation on bot-generated content that can help us to understand the role of impersonators in the production of fake content on Instagram.

cs.SI↗

arXiv2020

Hate Speech Detection and Racial Bias Mitigation in Social Media based on BERT model

Disparate biases associated with datasets and trained classifiers in hateful and abusive content identification tasks have raised many concerns recently. Although the problem of biased datasets on abusive language detection has been addressed more frequently, biases arising from trained classifiers have not yet been a matter of concern. Here, we first introduce a transfer learning approach for hate speech detection based on an existing pre-trained language model called BERT and evaluate the proposed model on two publicly available datasets annotated for racism, sexism, hate or offensive content on Twitter. Next, we introduce a bias alleviation mechanism in hate speech detection task to mitigate the effect of bias in training set during the fine-tuning of our pre-trained BERT-based model. Toward that end, we use an existing regularization method to reweight input samples, thereby decreasing the effects of high correlated training set' s n-grams with class labels, and then fine-tune our pre-trained BERT-based model with the new re-weighted samples. To evaluate our bias alleviation mechanism, we employ a cross-domain approach in which we use the trained classifiers on the aforementioned datasets to predict the labels of two new datasets from Twitter, AAE-aligned and White-aligned groups, which indicate tweets written in African-American English (AAE) and Standard American English (SAE) respectively. The results show the existence of systematic racial bias in trained classifiers as they tend to assign tweets written in AAE from AAE-aligned group to negative classes such as racism, sexism, hate, and offensive more often than tweets written in SAE from White-aligned. However, the racial bias in our classifiers reduces significantly after our bias alleviation mechanism is incorporated. This work could institute the first step towards debiasing hate speech and abusive language detection systems.

cs.SI↗

arXiv2020

A First Instagram Dataset on COVID-19

The novel coronavirus (COVID-19) pandemic outbreak is drastically shaping and reshaping many aspects of our life, with a huge impact on our social life. In this era of lockdown policies in most of the major cities around the world, we see a huge increase in people and professional engagement in social media. Social media is playing an important role in news propagation as well as keeping people in contact. At the same time, this source is both a blessing and a curse as the coronavirus infodemic has become a major concern, and is already a topic that needs special attention and further research. In this paper, we provide a multilingual coronavirus (COVID-19) Instagram dataset that we have been continuously collected since March 30, 2020. We are making our dataset available to the research community at Github. We believe that this contribution will help the community to better understand the dynamics behind this phenomenon in Instagram, as one of the major social media. This dataset could also help study the propagation of misinformation related to this outbreak.

cs.SI↗

arXiv2020

How Impersonators Exploit Instagram to Generate Fake Engagement?

Impersonators on Online Social Networks such as Instagram are playing an important role in the propagation of the content. These entities are the type of nefarious fake accounts that intend to disguise a legitimate account by making similar profiles. In addition to having impersonated profiles, we observed a considerable engagement from these entities to the published posts of verified accounts. Toward that end, we concentrate on the engagement of impersonators in terms of active and passive engagements which is studied in three major communities including ``Politician'', ``News agency'', and ``Sports star'' on Instagram. Inside each community, four verified accounts have been selected. Based on the implemented approach in our previous studies, we have collected 4.8K comments, and 2.6K likes across 566 posts created from 3.8K impersonators during 7 months. Our study shed light into this interesting phenomena and provides a surprising observation that can help us to understand better how impersonators engaging themselves inside Instagram in terms of writing Comments and leaving Likes.

cs.SI↗

arXiv2019

A BERT-Based Transfer Learning Approach for Hate Speech Detection in Online Social Media

Generated hateful and toxic content by a portion of users in social media is a rising phenomenon that motivated researchers to dedicate substantial efforts to the challenging direction of hateful content identification. We not only need an efficient automatic hate speech detection model based on advanced machine learning and natural language processing, but also a sufficiently large amount of annotated data to train a model. The lack of a sufficient amount of labelled hate speech data, along with the existing biases, has been the main issue in this domain of research. To address these needs, in this study we introduce a novel transfer learning approach based on an existing pre-trained language model called BERT (Bidirectional Encoder Representations from Transformers). More specifically, we investigate the ability of BERT at capturing hateful context within social media content by using new fine-tuning methods based on transfer learning. To evaluate our proposed approach, we use two publicly available datasets that have been annotated for racism, sexism, hate, or offensive content on Twitter. The results show that our solution obtains considerable performance on these datasets in terms of precision and recall in comparison to existing approaches. Consequently, our model can capture some biases in data annotation and collection process and can potentially lead us to a more accurate model.

cs.SI↗

arXiv2019

Uncovering Flaming Events on News Media in Social Media

Social networking sites (SNSs) facilitate the sharing of ideas and information through different types of feedback including publishing posts, leaving comments and other type of reactions. However, some comments or feedback on SNSs are inconsiderate and offensive, and sometimes this type of feedback has a very negative effect on a target user. The phenomenon known as flaming goes hand-in-hand with this type of posting that can trigger almost instantly on SNSs. Most popular users such as celebrities, politicians and news media are the major victims of the flaming behaviors and so detecting these types of events will be useful and appreciated. Flaming event can be monitored and identified by analyzing negative comments received on a post. Thus, our main objective of this study is to identify a way to detect flaming events in SNS using a sentiment prediction method. We use a deep Neural Network (NN) model that can identity sentiments of variable length sentences and classifies the sentiment of SNSs content (both comments and posts) to discover flaming events. Our deep NN model uses Word2Vec and FastText word embedding methods as its training to explore which method is the most appropriate. The labeled dataset for training the deep NN is generated using an enhanced lexicon based approach. Our deep NN model classifies the sentiment of a sentence into five classes: Very Positive, Positive, Neutral, Negative and Very Negative. To detect flaming incidents, we focus only on the comments classified into the Negative and Very Negative classes. As a use-case, we try to explore the flaming phenomena in the news media domain and therefore we focused on news items posted by three popular news media on Facebook (BBCNews, CNN and FoxNews) to train and test the model.

cs.SI↗

arXiv2019

Internet of Things-aided Smart Grid: Technologies, Architectures, Applications, Prototypes, and Future Research Directions

Traditional power grids are being transformed into Smart Grids (SGs) to address the issues in existing power system due to uni-directional information flow, energy wastage, growing energy demand, reliability and security. SGs offer bi-directional energy flow between service providers and consumers, involving power generation, transmission, distribution and utilization systems. SGs employ various devices for the monitoring, analysis and control of the grid, deployed at power plants, distribution centers and in consumers' premises in a very large number. Hence, an SG requires connectivity, automation and the tracking of such devices. This is achieved with the help of Internet of Things (IoT). IoT helps SG systems to support various network functions throughout the generation, transmission, distribution and consumption of energy by incorporating IoT devices (such as sensors, actuators and smart meters), as well as by providing the connectivity, automation and tracking for such devices. In this paper, we provide a comprehensive survey on IoT-aided SG systems, which includes the existing architectures, applications and prototypes of IoT-aided SG systems. This survey also highlights the open issues, challenges and future research directions for IoT-aided SG systems.

cs.NI↗

arXiv2018

Inspecting Interactions: Online News Media Synergies in Social Media

The rising popularity of social media has radically changed the way news content is propagated, including interactive attempts with new dimensions. To date, traditional news media such as newspapers, television and radio have already adapted their activities to the online news media by utilizing social media, blogs, websites etc. This paper provides some insight into the social media presence of worldwide popular news media outlets. Despite the fact that these large news media propagate content via social media environments to a large extent and very little is known about the news item producers, providers and consumers in the news media community in social media.To better understand these interactions, this work aims to analyze news items in two large social media, Twitter and Facebook. Towards that end, we collected all published posts on Twitter and Facebook from 48 news media to perform descriptive and predictive analyses using the dataset of 152K tweets and 80K Facebook posts. We explored a set of news media that originate content by themselves in social media, those who distribute their news items to other news media and those who consume news content from other news media and/or share replicas. We propose a predictive model to increase news media popularity among readers based on the number of posts, number of followers and number of interactions performed within the news media community. The results manifested that, news media should disperse their own content and they should publish first in social media in order to become a popular news media and receive more attractions to their news items from news readers.

cs.SI↗

arXiv2018

Electric Vehicle Charge Scheduling Mechanism to Maximize Cost Efficiency and User Convenience

This paper investigates the fee scheduling problem of electric vehicles (EVs) at the micro-grid scale. This problem contains a set of charging stations controlled by a central aggregator. One of the main stakeholders is the operator of the charging stations, who is motivated to minimize the cost incurred by the charging stations, while the other major stakeholders are vehicle owners who are mostly interested in user convenience, as they want their EVs to be fully charged as soon as possible. A bi-objective optimization problem is formulated to jointly optimize two factors that correspond to these stakeholders. An online centralized scheduling algorithm is proposed and proven to provide a Pareto-optimal solution. Moreover, a novel low-complexity distributed algorithm is proposed to reduce both the transmission data rate and the computation complexity in the system. The algorithms are evaluated through simulation, and results reveal that the charging time in the proposed method is 30% less than that of the compared methods proposed in the literature. The data transmitted by the distributed algorithm is 33:25% lower than that of a centralized one. While the performance difference between the centralized and distributed algorithms is only 2%, the computation time shows a significant reduction.

math.OC↗

arXiv2018

CPVNF:Cost-efficient Proactive VNF Placement and Chaining for Value-Added Services in Content Delivery Networks

Value-added services (e.g., overlaid video advertisements) have become an integral part of today's Content Delivery Networks (CDNs). To offer cost-efficient, scalable and more agile provisioning of new value-added services in CDNs, Network Functions Virtualization (NFV) paradigm may be leveraged to allow implementation of fine-grained services as a chain of Virtual Network Functions (VNFs) to be placed in CDN. The manner in which these chains are placed is critical as it both affects the quality of service (QoS) and provider cost. The problem is however, very challenging due to the specifics of the chains (e.g.,one of their end-points is not known prior to the placement). We formulate it as an Integer Linear Program (ILP) and propose a cost efficient Proactive VNF placement and chaining (CPVNF)algorithm. The objective is to find the optimal number of VNFs along with their locations in such a manner that the cost is minimized while QoS is met. Apart from cost minimization, the support for large-scale CDNs with a large number of servers and end-users is an important feature of the proposed algorithm. Through simulations, the algorithm's behaviour for small-scale to large-scale CDN networks is analyzed.

cs.NI↗

arXiv2017

When Traffic Flow Prediction Meets Wireless Big Data Analytics

Traffic flow prediction is an important research issue for solving the traffic congestion problem in an Intelligent Transportation System (ITS). Traffic congestion is one of the most serious problems in a city, which can be predicted in advance by analyzing traffic flow patterns. Such prediction is possible by analyzing the real-time transportation data from correlative roads and vehicles. This article first gives a brief introduction to the transportation data, and surveys the state-of-the-art prediction methods. Then, we verify whether or not the prediction performance is able to be improved by fitting actual data to optimize the parameters of the prediction model which is used to predict the traffic flow. Such verification is conducted by comparing the optimized time series prediction model with the normal time series prediction model. This means that in the era of big data, accurate use of the data becomes the focus of studying the traffic flow prediction to solve the congestion problem. Finally, experimental results of a case study are provided to verify the existence of such performance improvement, while the research challenges of this data-analytics-based prediction are presented and discussed.

cs.AI↗

arXiv2017

Getting Virtualized Wireless Sensor Networks IaaS Ready for PaaS

With the recent advances in sensor hardware and software, architectures for virtualized Wireless Sensor Networks (vWSNs) are now emerging. Through node- and network-level virtualization, vWSNs can be offered as Infrastructure-as-a-Service (IaaS) which can aid in realizing the true potential of Internet-of-Things (IoT). Cloud computing offers elastic provisioning of large-scale infrastructures to multiple concurrent users where Platform-as-a-Service (PaaS) interacts with IaaS in order to efficiently host and execute applications over these infrastructures. Amalgamating IoT with cloud computing potentially allows rapid application and service provisioning in an efficient, scalable and robust manner. However, interactions between vWSNs and PaaS are largely an unexplored area. Indeed, existing vWSN IaaS are not yet ready for PaaS. This paper proposes a vWSN IaaS architecture which is ready for interactions with PaaS. The proposed architecture is based on our previous works and is rooted in the fundamental differences between traditional IaaS and vWSN IaaS. We built a prototype using Java Sunspot as the WSN tool kit and made early performance measurements.

cs.NI↗