Detecting spam web pages through content analysis

Ntoulas, Alexandros; Najork, Marc; Manasse, Mark S.; Fetterly, Dennis

doi:10.1145/1135777.1135794

Cited by 469 publications

(396 citation statements)

References 9 publications

Supporting

Mentioning

355

Contrasting

Unclassified

Order By: Relevance

“…It has been traditionally studied in the context of e-mail [2], and Web pages [5,15]. The detection of opinion spam, i.e., the identification of fake reviews that try to deliberately mislead human readers, is just another face of the same problem [18].…”

Section: Related Workmentioning

confidence: 99%

Detection of Opinion Spam with Character n-grams

Fusilier

Montes-y-Gómez

Rosso

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper we consider the detection of opinion spam as a stylistic classification task because, given a particular domain, the deceptive and truthful opinions are similar in content but differ in the way opinions are written (style). Particularly, we propose using character ngrams as features since they have shown to capture lexical content as well as stylistic information. We evaluated our approach on a standard corpus composed of 1600 hotel reviews, considering positive and negative reviews. We compared the results obtained with character n-grams against the ones with word n-grams. Moreover, we evaluated the effectiveness of character n-grams decreasing the training set size in order to simulate real training conditions. The results obtained show that character n-grams are good features for the detection of opinion spam; they seem to be able to capture better than word n-grams the content of deceptive opinions and the writing style of the deceiver. In particular, results show an improvement of 2.3% and 2.1% over the word-based representations in the detection of positive and negative deceptive opinions respectively. Furthermore, character n-grams allow to obtain a good performance also with a very small training corpus. Using only 25% of the training set, a Naïve Bayes classifier showed F1 values up to 0.80 for both opinion polarities.

show abstract

Section: Related Workmentioning

confidence: 99%

Detection of Opinion Spam with Character n-grams

Fusilier

Montes-y-Gómez

Rosso

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…To counter this, researchers have proposed ways to track concept drift [8], which spammers use to thwart frequency-based content analysis. Comprehensive studies on content analysis have been proposed both for spam [30] and phishing sites [46,50]. Google is also performing phishing detection through content analysis [44], and researchers have used the search engine's index to identify scams campaigns with similar content [16].…”

Section: Related Workmentioning

confidence: 99%

Eyes of a Human, Eyes of a Program: Leveraging Different Views of the Web for Analysis and Detection

Corbetta

Invernizzi

Kruegel

et al. 2014

Research in Attacks, Intrusions and Defenses

View full text Add to dashboard Cite

Abstract. With JavaScript and images at their disposal, web authors can create content that is immediately understandable to a person, but is beyond the direct analysis capability of computer programs, including security tools. Conversely, information can be deceiving for humans even if unable to fool a program. In this paper, we explore the discrepancies between user perception and program perception, using content obfuscation and counterfeit "seal" images as two simple but representative case studies. In a dataset of 149,700 pages we found that benign pages rarely engage in these practices, while uncovering hundreds of malicious pages that would be missed by traditional malware detectors. We envision that this type of heuristics could be a valuable addition to existing detection systems. To show this, we have implemented a proofof-concept detector that, based solely on a similarity score computed on our metrics, can already achieve a high precision (95%) and a good recall (73%).

show abstract

“…Unfortunately, this also results in spamming, which refers to human activities that deliberately mislead search engines to rank some pages higher than they deserve. The following description of web spam taxonomy is basedon[l], [2], [5]and [12].…”

Section: Web Spam and Ranking Problemmentioning

confidence: 99%

“…So, in our opinion, web spam detection can not be simply considered as a problem of classification which most of the traditional work do [2,4]. In fact, it can be regarded as a ranking problem which arises recently in the social science and in information retrieval where human preferences play a major role [13,14].…”

Section: Web Spam and Ranking Problemmentioning

confidence: 99%

“…Second, if a user searches for information that is relevant to your pages but your pages are ranked low by search engines, then the user may not see the pages because one seldom clicks a large number of returned pages. Finally, a search engine may waste significant resources on spam pages because spam pages consume crawling bandwidth, pollute the web, and distort search ranking [2].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Content Trust Model for Detecting Web Spam

Wang

Zeng

IFIP International Federation for Information Processing

View full text Add to dashboard Cite

Abstract. As it gets easier to add information to the web via html pages, wikis, blogs, and other documents, it gets tougher to distinguish accurate or trustworthy information from inaccurate or untrustworthy information. Moreover, apart from inaccurate or untrustworthy information, we also need to anticipate web spam -where spammers publish false facts and scams to deliberately mislead users. Creating an effective spam detection method is a challenge. In this paper, we use the notion of content trust for spam detection, and regard it as a ranking problem. Evidence is utilized to define the feature of spam web pages, and machine learning techniques are employed to combine the evidence to create a highly efficient and reasonably-accurate spam detection algorithm. Experiments on real web data are carried out, which show the proposed method performs very well in practice.

show abstract

Detecting spam web pages through content analysis

Cited by 469 publications

References 9 publications

Detection of Opinion Spam with Character n-grams

Detection of Opinion Spam with Character n-grams

Eyes of a Human, Eyes of a Program: Leveraging Different Views of the Web for Analysis and Detection

Content Trust Model for Detecting Web Spam

Contact Info

Product

Resources

About