Online Social Networks are perpetually evolving and used in plenteous applications such as content sharing, chatting, making friends/followers, customer engagements, commercials, product reviews/promotions, online games, and news, etc. The substantial issues related to the colossal flood of social spam in social media are polarizing sentiments, impacting users’ online interaction time, degrading available information quality, network bandwidth, computing power, and speed. Simultaneously, groups of coordinated automated accounts/bots often use social networking sites to spread spam, rumors, bogus reviews, and fake news for targeted users or mass communication. The latest developments in the form of artificial intelligence-enabled Deepfakes have exacerbated these issues at large. Consequently, it becomes extremely relevant to review recent work concerning social spam and spammer detection to counter this issue and its effect. This paper provides a brief introduction to social spam, the spamming process, and social spam taxonomy. The comprehensive review entails several dimensionality reduction techniques used for feature selection/extraction, features used, various machine learning and deep learning techniques used for social spam and spammer detection, and their merits and demerits. Artificial intelligence and deep learning empowered Deepfake (text, image, and video) spam, and their countermeasures are also explored. Furthermore, meticulous discussions, existing challenges, and emerging issues such as robustness of detection systems, scalability, real-time datasets, evade strategies used by spammers, coordinated inauthentic behavior, and adversarial attacks on machine learning-based spam detectors, etc., have been discussed with possible directions for future research.

A review on social spam detection: Challenges, open issues, and future directions

Feature Selection has been a significant preprocessing procedure for classification in the area of Supervised Machine Learning. It is mostly applied when the attribute set is very large. The large set of attributes often tend to misguide the classifier. Extensive research has been performed to increase the efficacy of the predictor by finding the optimal set of features. The feature subset should be such that it enhances the classification accuracy by the removal of redundant features. We propose a new feature selection mechanism, an amalgamation of the filter and the wrapper techniques by taking into consideration the benefits of both the methods. Our hybrid model is based on a two phase process where we rank the features and then choose the best subset of features based on the ranking. We validated our model with various datasets, using multiple evaluation metrics. Furthermore, we have also compared and analyzed our results with previous works. The proposed model outperformed many existent algorithms and has given us good results.

/pdf/mini-batch-normalized-mutual-information-a-hybrid-feature-40w946ww8s.pdf

Mini-Batch Normalized Mutual Information: A Hybrid Feature Selection Method

Click Fraud is a fraudulent act of clicking on pay-per-click advertisements to increase the site’s revenue or to drain revenue from the advertiser. This illegal act has been putting commercial industries in a dilemma for quite some time. These industries think twice before advertising their products on websites and mobile-apps, as many parties try to exploit them. To safely promote their products, there must be an efficient system to detect click fraud. To address this problem, we propose a model called CFXGB (Cascaded Forest and XGBoost). The proposed model, classified under supervised machine learning, is a combination of two learning models used for feature transformation and classification. We showcase its superior performance compared to other related models, and make a comparison with multiple click fraud datasets with varying sizes.

A hybrid and effective learning approach for Click Fraud detection

Click fraud refers to the practice of generating random clicks on a link in order to extract illegitimate revenue from the advertisers. We present a generalized model for modeling temporal click fraud data in the form of probability or learning based anomaly detection and time series modeling with time scales like minutes and hours. The proposed approach consists of seven stages: Pre-processing, data smoothing, fraudulent pattern identification, homogenizing variance, normalizing auto-correlation, developing the AR and MA models and fine tuning along with evaluation of the models. The objective of the proposed work is to first, model multi-time-scale time series data on AR/MA by relying only on time and the label without the need of too many attributes and secondly, to model different time scales separately on Auto-regression (AR) and Moving Average (MA) models. Then, we evaluate the models by tuning forecasting errors and also by minimizing Akaike Information Criteria (AIC) and Bayesian Information Criteria (BIC) to obtain a best fit model for all time scale data. Through our experiments we also demonstrated that the Probability based model approach is better as compared to the Learning based probabilistic estimator model.

A Multi-time-scale Time Series Analysis for Click Fraud Forecasting using Binary Labeled Imbalanced Dataset

Online social networks (OSN) are one of the favorite places where people share posts like their photos, videos, and text to gain popularity. On the other hand, the marketing industry tries to gain the popularity of their advertisement using such OSNs. Popularity of a particular post depends on the number of likes received by that post. To increase one’s social worth, people try to use this market by artificially increasing the likes on their posts. There is a lack of research in the current literature on Instagram which is one of the growing OSNs. Our work focuses on detecting valid and fake like of posts with the application of learning model taking into consideration several popular factors. We developed an automated learning model to detect fake liking behavior on the Instagram post. The learned model can accurately differentiate between the legitimate and fake liker with an accuracy of 97% with ensemble-based learning model and also autoencoder is used to detect bots activity.

Learning-Based Model to Fight against Fake Like Clicks on Instagram Posts

Click fraud is a fast-growing cyber-criminal activity with the aim of deceptively clicking on the advertisements to make the profit to the publisher or cause loss to the advertiser. Due to the popularity of smartphones since the last decade, most of the modern-day advertisement businesses have been shifting their focus toward mobile platforms. Nowadays, in-app advertisement on mobile platforms is the most targeted victim of click fraud. Malicious entities launch attacks by clicking ads to artificially increase the click rates of specific ads without the intention of using them for legitimate purposes. The fraud clicks are supposed to be caught by the ad providers as part of their service to the advertisers; however, there is a lack of research in the current literature for addressing and evaluating different techniques of click fraud detection and prevention. Another challenge toward click fraud detection is that the attack model can itself be an active learning system (smart attacker) with the aim of actively misleading the training process of fraud detection model via polluting the training data. In this paper, we propose a deep-learning based model to address the challenges as mentioned above. The model is a hybrid of artificial neural network (ANN), autoencoder and semi-supervised generative adversarial network (GAN). Our proposed approach triumphs excellent accuracy than other models.

https://dl.acm.org/doi/pdf/10.1145/3299815.3314453

Deep Learning-based Model to Fight Against Ad Click Fraud

Lung cancer is one of the commonest diseases globally and accounts for more deaths from lung cancer than other ones. We did an exact review of the distributed things on the study of disease transmission, conclusion, and treatment of lung cancer and as the dispersion of lung cancer is rapidly going on, it reaches the epidemic range or peak level. Besides, it has beaten the earlier commonest outline of cancer and is presently the commonest harm in men and women. Different modalities for early recognition through screening are being researched and it is observed that the majority of the patients have died due to late detection of this disease so recognition at the premature stage, as well as accurate detection of a lung nodule, is very much necessary so the most challenging thing is to develop a robust method for nodule detection that is why the main objective of this paper is early detection, minimizing the false rate, as well as improving accuracy. In the paper, a new model is proposed based on deep-learning for the pinpointing of region precisely, as we know there is a huge advancement in Deep convolutional neural networks in pre-processing, we adopted Three-dimensional Convolutional Neural Network) for the prognosis of lung cancer from the CT images of the long-sufferer. As one knows about the convolution neural network, it makes things easier to extract significant data by processing the images of the dataset. Gradient-weighted class activation mapping strategies can convey the information by visualization for precise judgment in lung knot by point-up the adversely affected area. 

Lung carcinoma detection at premature stage using deep learning techniques

AIM
To study the clinico-etiological profile of children with thrombocytopeniaMethods: This prospective hospital-based study included all children (<18 years) with thrombocytopenia at the time of hospitalization and/or thrombocytopenia during the course of their hospital stay. A detailed history was recorded and appropriate laboratory investigations were carried out Results: The study group comprised 246 children (mean age, 9.29 years; median age, 10 years) with male to female ratio of 1.5:1. Nearly 45% of children were above 10 years of age. Trends of admissions showed that the majority of children with thrombocytopenia (n = 115) got hospitalized during the rainy season, followed by summer (n = 84). Fever (72.8%), pallor (52.8%), bleeding manifestations (22%), lymphadenopathy (20.3%), and splenomegaly (20.3%) were common clinical features. Petechiae was the most common bleeding manifestation (63%). Septicemia (24%) was the most common etiology, followed by megaloblastic anemia (14.6%), undiagnosed fever (10.2%), local infection (9.3%), hepatitis (6.5%), and scrub typhus (6.1%). About nine children died. All those who died had septicemia and multi-organ dysfunction (MOD). On logistic regression analysis, age >10 years, presence of bleeding, arthralgia, rash, pallor, gastrointestinal (GI) symptoms, hematological disorders, and malignancy were associated with severe thrombocytopeniaConclusion: Thrombocytopenia is a common hematological observation. This study revealed seasonal variation in the occurrence of thrombocytopenia in children, with the maximum number of cases in the rainy season. Septicemia is the commonest etiology. The majority of children with thrombocytopenia have no bleeding manifestations. Age >10 years, presence of bleeding, arthralgia, rash, pallor, GI symptoms, hematological disorders, and malignancy are associated with severe thrombocytopenia.

/pdf/thrombocytopenia-in-children-a-large-prospective-study-on-378n7jl9.pdf

Isha Bhatia

Papers

Deep Learning-based Model to Fight Against Ad Click Fraud

Lung carcinoma detection at premature stage using deep learning techniques

Thrombocytopenia in Children: A Large Prospective Study on Clinical Manifestations, Seasonal Variation, Etiology, and Outcome.