NeuroinformaticsNatural LanguageProcessing (NeuroNLP) relies on clustering and classification for information categorization of biologically relevant extraction targets and for interconnections to knowledge-related patterns in event and text mined datasets. The accuracy of machine learning algorithms depended on quality of text-mined data while efficacy relied on the context of the choice of techniques. Although developments of automated keyword extraction methods have made differences in the quality of data selection, the efficacy of the Natural Language Processing (NLP) methods using verified keywords remain a challenge. In this paper, we studied the role of text classification and document clustering algorithms on datasets, where features were obtained by mapping to manually verified MESH terms published by National Library of Medicine (NLM). In this study, NLP data classification involved comparing Stechniques and unsupervised learning was performed with 6 clustering algorithms. Most classification techniques except meta-based algorithms namely stacking and vote, allowed 90% or higher training accuracy. Test accuracy was high (=>95%) probably due to limited test dataset. Logistic Model Trees had 30-fold higher runtime compared to other classification algorithms including Naive Bayes, AdaBoost, Hoeffding Tree. Grouped error rate in clustering was 0-4%. Runtime-wise, clustering was faster than classification algorithms on MESH-mapped NLP data suggesting clustering methods as adequate towards Medline-related datasets and text-mining big data analytic systems.

No abstract

Abstract-With many online engineering platforms such as virtual and remote laboratories designed for young or aged users, user authentication and passwords-based methods are being re-evaluated for tracking usage patterns and security. For ICT-enabled online engineering platforms, image-based humancentric approaches are gaining relevance for access frameworks. With the rubber-hose attacks, increased senior users, many existing systems are vulnerable to many attacks. This paper employs human uniqueness of narrative skills on an image-based password system for online platforms with focus on theme in the password generation process. To generate the secret password, a specially designed computer game was used. We used narrative constructs composed of cartoon image sequences to generate user-speci!c secret key. The durability of generated passwords and the authentication process while assessing the reconstruction process by a potential hacker was verified. For validating use of coerced attacks, under imposed psychological duress, users failed retrieving the password sequence suggesting the reliability as an anti-coercive attack cybersecurity tool. A set of experiments were used to analyze user behavior behind the image-based password system. EEG measurements demonstrated increased activity of " rhythms in F3 and FC5 channel bins and augmented levels of # rhythms in F3 and O1 channels, suggesting users added personalization to authentication more than in alpha-numeric password-based logins.

No abstract

Abstract. Existing cryptographic systems use strong passwords but several techniques are vulnerable to rubber-hose attacks, wherein the user is forced to reveal the secret key. This paper specifies a defense technique against rubberhose attacks by taking advantage of image sequence-based theme selection, dependent on a user's personal construct and active implicit learning. In this paper, an attempt to allow the human brain to generate the password via a computer task of arranging themed images through which the user learns a password without any conscious knowledge of the learned pattern. Although used in authentication, users cannot be coerced into revealing the secret key since the user has no direct knowledge on the choice of the learned secret. We also show that theme interception sequence learning tool works significantly well with mixed user age groups and can be used as a secondary layer of security where human user authentication remains a priority.