Spam Identification Research Articles

Deep Learning Algorithm Models for Spam Identification on Cellular Short Message Service

Nowadays, the types and products of cellular telecommunications services are very diverse, especially with the upcoming of 5G technology, which makes telecommunications service products such as voice, video, and text messages rely on data packages. Even though the digital era is rapidly growing, the Short Messaging Service (SMS) is still relevant and used as a telecommunication service despite so many sophisticated instant messaging services that rely on the internet. Smartphone users especially in Indonesia are often terrorized by spam messages with pretentious content. Moreover, the SMS came from an unknown number and contained a message or link to a fraudulent site. This study develops a Deep Learning model to predict whether a short text message (SMS) is important or spam. This research domain belongs to Natural Language Processing (NLP) for text processing. The models used are Dense Network, Long Short Term Memory (LSTM), and Bi-directional Long Short Term Memory (Bi-LSTM). Based on the evaluation of the Dense Network model, it produces a loss of 14.22% and an accuracy of 95.63%. The evaluation of the LSTM model is 19.89% loss and 94.76% accuracy. Finally, the evaluation of the Bi-LSTM model is 19.88% loss and 94.75% accuracy.

An Improved Framework for Content- and Link-Based Web-Spam Detection: A Combined Approach

In this modern era, people utilise the web to share information and to deliver services and products. The information seekers use different search engines (SEs) such as Google, Bing, and Yahoo as tools to search for products, services, and information. However, web spamming is one of the most significant issues encountered by SEs because it dramatically affects the quality of SE results. Web spamming’s economic impact is enormous because web spammers index massive free advertising data on SEs to increase the volume of web traffic on a targeted website. Spammers trick an SE into ranking irrelevant web pages higher than relevant web pages in the search engine results pages (SERPs) using different web-spamming techniques. Consequently, these high-ranked unrelated web pages contain insufficient or inappropriate information for the user. To detect the spam web pages, several researchers from industry and academia are working. No efficient technique that is capable of catching all spam web pages on the World Wide Web (WWW) has been presented yet. This research is an attempt to propose an improved framework for content- and link-based web-spam identification. The framework uses stopwords, keywords’ frequency, part of speech (POS) ratio, spam keywords database, and copied-content algorithms for content-based web-spam detection. For link-based web-spam detection, we initially exposed the relationship network behind the link-based web spamming and then used the paid-link database, neighbour pages, spam signals, and link-farm algorithms. Finally, we combined all the content- and link-based spam identification algorithms to identify both types of spam. To conduct experiments and to obtain threshold values, WEBSPAM-UK2006 and WEBSPAM-UK2007 datasets were used. A promising F-measure of 79.6% with 81.2% precision shows the applicability and effectiveness of the proposed approach.

Classifiers selection based on analytic hierarchy process and similarity score for spam identification

The overrun of images and corpus spam on the web constitutes a threat to users producing more troubles, diminishing internet bandwidth, and increasing network overload. In this paper, we introduce a multi-criteria decision-making (MCDM) framework for spam classification based on an analytic hierarchy process. The proposed approach permits the selection of the most optimal classifier among six machine learning algorithms according to a set of performance metrics. The ranking consistency of classifiers has been evaluated with a consistency ratio almost inferior to 0.1 on all dataset. The proposed framework can recommend top-K classifiers with high accuracy by using Euclidean distance on basic characteristics of test and train data that was already trained on a set of classifiers and fed into the MCDM framework. The experiment has been performed on eight publicly available datasets and tested on four datasets using only known features from the dataset and using Euclidean distance measure to identify from the most performant classifier algorithm. The followed methodology has accurately identified the most suited classifier on training datasets with average accuracy 95% for both images and corpus spam out and achieved 100% accuracy on a personalimageham+spamarchive_jml dataset that performs better than existing methods. We performed Friedman statistical test on the dataset that showed significance, then a post hoc Conover test performed on each data to make a pairwise comparison that demonstrated the effectiveness of the proposed approach based AHP framework. The proposed approach outperforms existing research works in term of spam identification on both image and corpus spam data and was successful in recommending classifiers to unknown datasets.

A Study on Opinion Spamming: Fake Consumer Review Detection

Online audits are the most important wellsprings of data about client feelings and are considered the columns on which the standing of an association is assembled. From a client's viewpoint, audit data is vital to settle on an appropriate choice with respect to an online buy. Surveys are for the most part thought to be a fair-minded assessment of a person's very own involvement in an item, however, the fundamental truth about these audits recounts an alternate story. Spammers abuse these audit stages unlawfully on account of impetuses engaged with composing counterfeit surveys, subsequently attempting to acquire a bit of leeway over contenders bringing about an unstable development of assessment spamming. This training is known as Opinion (Review) Spam, where spammers control and toxic substance surveys (i.e., making phony, untruthful, or misleading audits) for benefit or gain. It has become a typical practice for individuals to discover and to understand assessments/surveys on the Web for some reason. For instance, in the event that one needs to purchase an item, one commonly goes to a vendor or audit site (e.g., amazon.com) to peruse a few surveys of existing clients of the item. In the event that one sees numerous positive audits of the item, one is probably going to purchase the item. Notwithstanding, in the event that one sees many negative surveys, he/she will in all probability pick another item. Positive suppositions can bring about huge monetary benefits and additionally popularities for associations and people. This, sadly, offers great motivating forces for input spam. Most of the momentum research has zeroed in on regulated learning strategies, which require named information, a shortage with regards to online survey spam. Examination of techniques for Big Data is of revenue, since there are a huge number of online audits, with a lot seriously being produced every day. Until now, we have not discovered any papers that review the impacts of Big Data examination for survey spam identification. The essential objective of this paper is to give a solid and far-reaching similar investigation of flow research on identifying audit spam utilizing different AI procedures and to devise a strategy for directing further examination.

A fuzzy Dempster–Shafer classifier for detecting Web spams

The Web spam identification problem can be modeled as an instance of the conventional classification problem. Web spams aim at deceiving web crawlers by advertising certain Web pages through elevation of their page rankings superficially than their actual weights. Web spams are intended to produce fraudulent results of web search queries and degenerate the client’s experience by directing users to fake Web pages. We present a fuzzy evidence-based methodology for identifying Web spams by which the spamicity of web hosts is formulated as a reasoning problem in the presence of uncertainty. However, any classification task intrinsically suffers from incomplete or vague evidence and ambiguity in the class assignment based on evidence. In this work, we combine fuzzy reasoning as the decision maker for selecting the most suitable evidence in a multi-source Dempster–Shafer (DS) based classification algorithm. The introduced approach has the benefit of providing more reliable solution to detect spams without any prior information. The evidence theory offers flexible support that takes into account the multi-dimensional nature of implementation decisions. The experimental results show that the fuzzy reasoning in combination with DS theory, reduces the conflicts among evidence leading to enhanced classification results. The aim of this paper is to describe the potential of fuzzy reasoning and the Dempster–Shafer Theory (DST) as a decision model for the web spams classification problem.

Effective Filtering of Unsolicited Messages from Online Social Networks Using Spam Templates and Social Contexts

Online social networking sites have shown an unbelievable widening in the last decade. Spammers utilise social networking sites to unroll spam messages due to its fame and use various procedures to spread spam. Consequently, the identification of spam must be well fortified enough to detect unsolicited messages and deter spammers. Though various spam identification procedures are obtainable, to improve the accuracy for spam identification is inevitable. In this work, a method to detect unsolicited messages is proposed to recognise and avert spam messages. The social context parameters such as trust and strength as well as spam template matching are also considered along with basic classifiers for effective spam classification. The intercommunication factors between the users are used for strength calculation. Spam template generation is performed based on the majority merge operation on the spam messages during the training time, and spam templates comparison is performed with the incoming messages during the testing time. Trust value updation is performed after the message classification. Experimental results demonstrate that the proposed model with SVM-Polynomial Radial Basis kernel which provides better accuracy in spam classification and outperforms all the state-of-the-art methods.

Spamex: A Multi Array Methodology to Detect Computerized Spammers in Twitter social media

The Twitter interpersonal organizations give an approach to clients to keep up contact with others. Expanding of the informal organization's quality allows every one of them to accumulate monstrous measures of nonpublic information with respect to their clients. Tragically, this data riches just as its straightforwardness to get to clients' data could draw in malevolent gathering's consideration. That is the reason these systems have been attacked by spammers while, there have been a great deal of work to analyze and fix them. With reference to this issue spammers look for new ways that to concentrate on these systems every day, there have being nonstop activities to distinguish spammers and malignant email. The point of this paper is to take a gander at past works inside the field of spam identification in informal communities. In late time, on-line interpersonal organizations are loaded with differed undesirable dangers. Despite the fact that they gave us an open stage to impart our contemplations to other people, notwithstanding, because of abuse of this incredible asset, general clients are in jeopardized condition. For instance, YouTube has been utilized as a limited time ground by different craftsman to transfer their music recordings, film trailers, and so on and watchers can post their supposition on them. Tragically, regularly malevolent clients use to post phishing site connections, promotions, and deceitful data in the remarks segment, which may transmit infections or malwares.

Construction and Reduction Methods of Web Spam Identification Index System

Background: With the rapid development of the Internet, the number of web spam has increased dramatically in recent years, which has wasted search engine storage and computing power on a massive scale. To identify the web spam effectively, the content features, link features, hidden features and quality features of web page are integrated to establish the corresponding web spam identification index system. However, the index system is highly correlation dimension. Methods: An improved method of autoencoder named stacked autoencoder neural network (SAE) is used to realize the reduction of the web spam identification index system. Results: The experiment results show that our method could reduce effectively the index of web spam and significantly improves the recognition rate in the following work. Conclusion: An autoencoder based web spam indexes reduction method is proposed in this paper. The experimental results show that it greatly reduces the temporal and spatial complexity of the future web spam detection model.

Categorization of spam images and identification of controversial images on mobile phones using machine learning and predictive learning

The surge in popularity of instant messaging apps has greatly increased the volume of media exchanges, majorly consisting of images. This has resulted in cluttering of images, low storage, etc. In this paper, we are addressing these issues by proposing a system that combines various techniques of facial recognition, text extraction and Natural language processing and refines them through novel approaches to mark images as important or spam according to user’s behaviour and preferences. Our system classifies the images on user’s smartphone into separate categories and processes them further accordingly. Apart from the prominent categories based on the usability of the image, the system also recognizes duplicate and similar images which facilitates their easy deletion.

An automated text categorization framework based on hyperparameter optimization

A great variety of text tasks such as topic or spam identification, user profiling, and sentiment analysis can be posed as a supervised learning problem and tackled using a text classifier. A text classifier consists of several subprocesses, some of them are general enough to be applied to any supervised learning problem, whereas others are specifically designed to tackle a particular task using complex and computational expensive processes such as lemmatization, syntactic analysis, etc. Contrary to traditional approaches, we propose a minimalist and multi-propose text-classifier able to tackle tasks independently of domain and language. We named our approach μTC. Our approach is composed of several easy-to-implement text transformations, text representations, and a supervised learning algorithm. These pieces produce a competitive classifier in several challenging domains such as informally written text. We provide a detailed description of μTC along with an extensive experimental comparison with relevant state-of-the-art methods, i.e., μTC was compared on 30 different datasets. Regarding accuracy, μTC obtained the best performance in 20 datasets while achieves competitive results in the remaining ones. The compared datasets include several problems like topic and polarity classification, spam detection, user profiling and authorship attribution. Furthermore, our approach allows the usage of the technology even without an in-depth knowledge of machine learning and natural language processing.

Identifikasi Komentar Spam Pada Instagram

Spam pada Instagram (IG) umumnya berupa komentar yang dianggap mengganggu karena tidak berhubungan dengan foto atau video yang dikomentari. Spam pada komentar dapat menyebabkan beberapa dampak negatif seperti menyulitkan untuk mengikuti diskusi pada komentar yang dipenuhi oleh komentar spam dan menyebabkan seseorang tampak populer karena jumlah komentarnya banyak walaupun pada kenyataannya lebih banyak komentar yang berupa spam. Penelitian ini mencoba untuk membangun model yang dapat melakukan identifikasi komentar spam pada IG. Komentar pada IG berbentuk teks, sehingga pada penelitian ini digunakan metode-metode pengolahan teks. Untuk identifikasi digunakan metode Support Vector Machine (SVM). Data komentar yang digunakan pada penelitian ini dikumpulkan dari komentar-komentar pada foto atau video yang dibagikan oleh aktor dan artis Indonesia yang memiliki pengikut (follower) paling banyak di IG. Dari hasil penelitian didapatkan model identifikasi komentar spam dengan metode SVM menghasilkan tingkat akurasi 78,49% yang lebih baik jika dibandingkan dengan model pembanding yang menggunakan metode NB (77,25%). Penelitian ini juga menguji beberapa proporsi data pelatihan yang berbeda-beda dan hasilnya metode SVM tetap lebih baik dibandingkan dengan metode NB. Hasil lain dari penelitian ini adalah tahap pre-processing dan stemming yang harus disesuaikan terutama untuk dukungan terhadap pengolahan karakter-karakter unicode dan simbol-simbol khusus yang banyak ditemukan pada komentar-komentar di IG.

SMSAD: a framework for spam message and spam account detection

Short message communication media, such as mobile and microblogging social networks, have become attractive platforms for spammers to disseminate unsolicited contents. However, the traditional content-based methods for spam detection degraded in performance due to many factors. For instance, unlike the contents posted on social networks like Facebook and Renren, SMS and microblogging messages have limited size with the presence of many domain specific words, such as idioms and abbreviations. In addition, microblogging messages are very unstructured and noisy. These distinguished characteristics posed challenges to existing email spam detection models for effective spam identification in short message communication media. The state-of-the-art solutions for social spam accounts detection have faced different evasion tactics in the hands of intelligent spammers. In this paper, a unified framework is proposed for both spam message and spam account detection tasks. We utilized four datasets in this study, two of which are from SMS spam message domain and the remaining two from Twitter microblog. To identify a minimal number of features for spam account detection on Twitter, this paper studied bio-inspired evolutionary search method. Using evolutionary search algorithm, a compact model for spam account detection is proposed, which is incorporated in the machine learning phase of the unified framework. The results of the various experiments conducted indicate that the proposed framework is promising for detecting both spam message and spam account with a minimal number of features.

Optimizing Feedforward Neural Networks Using Biogeography Based Optimization for E-Mail Spam Identification

Spam e-mail has a significant negative impact on individuals and organizations, and is considered as a serious waste of resources, time and efforts. Spam detection is a complex and challenging task to solve. In literature, researchers and practitioners proposed numerous approaches for automatic e-mail spam detection. Learning-based filtering is one of the important approaches used for spam detection where a filter needs to be trained to extract the knowledge that can be used to detect the spam. In this context, Artificial Neural Networks is a widely used machine learning based filter. In this paper, we propose the use of a common type of Feedforward Neural Network called Multi-Layer Perceptron (MLP) for the purpose of e-mail spam identification, where the weights of this network model are found using a new nature-inspired metaheuristic algorithm called Biogeography Based Optimization (BBO). Experiments and results based on two different spam datasets show that the developed MLP model trained by BBO gets high generalization performance compared to other optimization methods used in the literature for e-mail spam detection.

Twitter spammer detection using data stream clustering

The rapid growth of Twitter has triggered a dramatic increase in spam volume and sophistication. The abuse of certain Twitter components such as “hashtags”, “mentions”, and shortened URLs enables spammers to operate efficiently. These same features, however, may be a key factor in identifying new spam accounts as shown in previous studies. Our study provides three novel contributions. Firstly, previous studies have approached spam detection as a classification problem, whereas we view it as an anomaly detection problem. Secondly, 95 one-gram features from tweet text were introduced alongside the user information analyzed in previous studies. Finally, to effectively handle the streaming nature of tweets, two stream clustering algorithms, StreamKM++ and DenStream, were modified to facilitate spam identification. Both algorithms clustered normal Twitter users, treating outliers as spammers. Each of these algorithms performed well individually, with StreamKM++ achieving 99% recall and a 6.4% false positive rate; and DenStream producing 99% recall and a 2.8% false positive rate. When used in conjunction, these algorithms reached 100% recall and a 2.2% false positive rate, meaning that our system was able to identify 100% of the spammers in our test while incorrectly detecting only 2.2% of normal users as spammers.

Comparison of Feature Vectors in Keystroke Dynamics

A number of current applications require algorithms able to extract a model from one-class data and classify unseen data as self or non-self in a novelty detection scenario, such as spam identification and intrusion detection. In this paper the authors focus on keystroke dynamics, which analyses the user typing rhythm to improve the reliability of user authentication process. However, several different features may be extracted from the typing data, making it difficult to define the feature vector. This problem is even more critical in a novelty detection scenario, when data from the negative class is not available. Based on a keystroke dynamics review, this work evaluated the most used features and evaluated which ones are more significant to differentiate a user from another using keystroke dynamics. In order to perform this evaluation, the authors tested the impact on two benchmark databases applying bio-inspired algorithms based on neural networks and artificial immune systems.

Bayesian Spam Filter Based on Distributed Architecture

The flood of spam promotes the development of anti-spam technology. In this paper, we bring forward the Bayesian filter technology based on the distributed architecture, which can realize the sharing of the Bayesian learning outcomes among servers within the system, so as to increase the accuracy of spam recognition. We, in the paper, discuss the sharing model of information with spam features under the distributed architecture and the spam identification process; analyze the Bayes algorithm and carry out the relevant improvements; design the Bayes Filter based on distributed architecture on the above basis and verify the effect of the filter by experiments.

An efficient approach to comment spam identification

This paper proposes a novel approach to comment spam identification based on content analysis. Three main features including the number of links, content repetitiveness, and text similarity are used for comment spam identification. In practice, content repetitiveness is determined by the length and frequency of the longest common substring. Furthermore, text similarity is calculated using vector space model. The precisions of preliminary experiments on comment spam identification conducted on Chinese and English are as high as 93% and 82% respectively. The results show the validity and language independency of this approach. Compared with conventional spam filtering approaches, our method requires no training, no rule sets and no link relationships. The proposed approach can also deal with new comments as well as existing comments.

An information granulation based data mining approach for classifying imbalanced data

Recently, the class imbalance problem has attracted much attention from researchers in the field of data mining. When learning from imbalanced data in which most examples are labeled as one class and only few belong to another class, traditional data mining approaches do not have a good ability to predict the crucial minority instances. Unfortunately, many real world data sets like health examination, inspection, credit fraud detection, spam identification and text mining all are faced with this situation. In this study, we present a novel model called the “Information Granulation Based Data Mining Approach” to tackle this problem. The proposed methodology, which imitates the human ability to process information, acquires knowledge from Information Granules rather then from numerical data. This method also introduces a Latent Semantic Indexing based feature extraction tool by using Singular Value Decomposition, to dramatically reduce the data dimensions. In addition, several data sets from the UCI Machine Learning Repository are employed to demonstrate the effectiveness of our method. Experimental results show that our method can significantly increase the ability of classifying imbalanced data.