Detecting Fake Medical Web Sites Using Recursive Trust Labeling

Abbasi, Ahmed; Zahedi, Fatemeh; Kaza, Siddharth

doi:10.1145/2382438.2382441

Cited by 54 publications

(23 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same problem exists in the domain of online health sites. To our knowledge, most studies of health-related trust have only tested individual specific points defined by a project [17] [18]. The lack of a de facto comprehensive standard for trust of online health information precludes comparison among different studies.…”

Section: Related Workmentioning

confidence: 99%

Feasibility of automated detection of HONcode conformity for health-related websites

Boyer¹,

Dolamic²

2014

IJACSA

View full text Add to dashboard Cite

Abstract-In this paper, authors evaluate machine learning algorithms to detect the trustworthiness of a website according to HONcode criteria of conduct (detailed in paper). To derive a baseline, we evaluated a Naive Bayes algorithm, using single words as features. We compared the baseline algorithm's performance to that of the same algorithm employing different feature types, and to the SVM algorithm. The results demonstrate that the most basic configuration (Naive Bayes, single word) could produce a 0.94 precision for "easy" HON criteria such as "Date". Conversely, for more difficult HON criteria "Justifiability", we obtained precision of 0.68 by adjusting the system parameters such as algorithm (SVM) and feature types (W2).

show abstract

Section: Related Workmentioning

confidence: 99%

Feasibility of automated detection of HONcode conformity for health-related websites

Boyer¹,

Dolamic²

2014

IJACSA

View full text Add to dashboard Cite

show abstract

“…An important area of future work includes the design of automated approaches to flagging inconsistent or unsubstantiated information [1]. In the present study, our goal was to "inform end users" [30] by exposing relevant website and document content features (described in Section 3) in the search result user interface, rather than attempt to predict credibility.…”

Section: Valuation Principlesmentioning

confidence: 99%

Patient-Centered Tools for Medication Information Search

Wilcox

Feiner

Elhadad

et al. 2014

Proceedings of the 8th International Conference on Pervasive Computing Technologies for Healthcare

View full text Add to dashboard Cite

Recent research focused on online health information seeking highlights a heavy reliance on general-purpose search engines. However, current general-purpose search interfaces do not necessarily provide adequate support for non-experts in identifying suitable sources of health information. Popular search engines have recently introduced search tools in their user interfaces for a range of topics. In this work, we explore how such tools can support non-expert, patientcentered health information search. Scoping the current work to medication-related search, we report on findings from a formative study focused on the design of patient-centered, medicationinformation search tools. Our study included qualitative interviews with patients, family members, and domain experts, as well as observations of their use of Remedy, a technology probe embodying a set of search tools. Post-operative cardiothoracic surgery patients and their visiting family members used the tools to find information about their hospital medications and were interviewed before and after their use. Domain experts conducted similar search tasks and provided qualitative feedback on their preferences and recommendations for designing these tools. Findings from our study suggest the importance of four valuation principles underlying our tools: credibility, readability, consumer perspective, and topical relevance.

show abstract

“…Bannur, Saul, and Savage (2011)s research has some similarities to this paper but it uses a very small dataset and furthermore this paper utilizes various other types of features. Abbasi, Zhang, Zimbra, Chen, and Nunamaker (2010) and Abbasi, Zahedi, and Kaza (2012) ran some classification to detect fake medical sites but the size of dataset was very small, whereas this paper focuses on detecting any type of malicious webpages. Fu, Wenyin, and Deng (2006) and Liu, Deng, Huang, and Fu (2006)s research considered the visual aspects of a webpage to determine whether the page is malicious or not.…”

Section: Introductionmentioning

confidence: 99%

Comparisons of machine learning techniques for detecting malicious webpages

Kazemian

Ahmed

2015

Expert Systems with Applications

View full text Add to dashboard Cite

a b s t r a c tThis paper compares machine learning techniques for detecting malicious webpages. The conventional method of detecting malicious webpages is going through the black list and checking whether the webpages are listed. Black list is a list of webpages which are classified as malicious from a user's point of view. These black lists are created by trusted organizations and volunteers. They are then used by modern web browsers such as Chrome, Firefox, Internet Explorer, etc. However, black list is ineffective because of the frequent-changing nature of webpages, growing numbers of webpages that pose scalability issues and the crawlers' inability to visit intranet webpages that require computer operators to log in as authenticated users. In this paper therefore alternative and novel approaches are used by applying machine learning algorithms to detect malicious webpages. In this paper three supervised machine learning techniques such as K-Nearest Neighbor, Support Vector Machine and Naive Bayes Classifier, and two unsupervised machine learning techniques such as K-Means and Affinity Propagation are employed. Please note that K-Means and Affinity Propagation have not been applied to detection of malicious webpages by other researchers. All these machine learning techniques have been used to build predictive models to analyze large number of malicious and safe webpages. These webpages were downloaded by a concurrent crawler taking advantage of gevent. The webpages were parsed and various features such as content, URL and screenshot of webpages were extracted to feed into the machine learning models. Computer simulation results have produced an accuracy of up to 98% for the supervised techniques and silhouette coefficient of close to 0.96 for the unsupervised techniques. These predictive models have been applied in a practical context whereby Google Chrome can harness the predictive capabilities of the classifiers that have the advantages of both the lightweight and the heavyweight classifiers.

show abstract

Detecting Fake Medical Web Sites Using Recursive Trust Labeling

Cited by 54 publications

References 48 publications

Feasibility of automated detection of HONcode conformity for health-related websites

Feasibility of automated detection of HONcode conformity for health-related websites

Patient-Centered Tools for Medication Information Search

Comparisons of machine learning techniques for detecting malicious webpages

Contact Info

Product

Resources

About