Neural networks and hybrid intelligent models: foundations, theory, and applications

h-Life: an intelligent personal health assistant

In this paper the architecture and subsystems of the h-Life platform are described. The system is currently under implementation and therefore all system components and modules are described according to their design specifications. h-Life aims at developing a user-friendly, personalized and ubiquitously accessible knowledge-based system that will provide citizens with an intelligent tool for monitoring the state of health and lifestyle. The h-Life system is based on the integration of several components: • a knowledge repository related to health, nutrition, fitness and lifestyle• a user repository where personal health records and user profiles are stored• a plan repository for the storage of medical plans• an intelligent adviser and plan selection mechanism, which provides advice and plans based on the user profile• an alerting and reminding module that issues alerts (to ensure the timely execution of a selected plan) and notifications concerning events of potential interest• a customer-facing module, which ensures personalized interaction with the system and customizes the user-system interaction according to the user profile.Based on the latest technology in intelligent and communication systems, the h-Life environment aims at providing the user with personalized, ‘certified’ lifestyle information and consultancy, tailored to individual needs and available anywhere and at anytime.

Comparing writing style feature-based classification methods for estimating user reputations in social media

In recent years, the anonymous nature of the Internet has made it difficult to detect manipulated user reputations in social media, as well as to ensure the qualities of users and their posts. To deal with this, this study designs and examines an automatic approach that adopts writing style features to estimate user reputations in social media. Under varying ways of defining Good and Bad classes of user reputations based on the collected data, it evaluates the classification performance of the state-of-art methods: four writing style features, i.e. lexical, syntactic, structural, and content-specific, and eight classification techniques, i.e. four base learners—C4.5, Neural Network (NN), Support Vector Machine (SVM), and Naïve Bayes (NB)—and four Random Subspace (RS) ensemble methods based on the four base learners. When South Korea’s Web forum, Daum Agora, was selected as a test bed, the experimental results show that the configuration of the full feature set containing content-specific features and RS-SVM combining RS and SVM gives the best accuracy for classification if the test bed poster reputations are segmented strictly into Good and Bad classes by portfolio approach. Pairwise t tests on accuracy confirm two expectations coming from the literature reviews: first, the feature set adding content-specific features outperform the others; second, ensemble learning methods are more viable than base learners. Moreover, among the four ways on defining the classes of user reputations, i.e. like, dislike, sum, and portfolio, the results show that the portfolio approach gives the highest accuracy.

Characterization of 14 novel deletions underlying Rubinstein-Taybi syndrome: an update of the CREBBP deletion repertoire

Rubinstein-Taybi syndrome (RSTS) is a rare, clinically heterogeneous disorder characterized by cognitive impairment and several multiple congenital anomalies. The syndrome is caused by almost private point mutations in the CREBBP (~55% of cases) and EP300 (~8%) genes. The CREBBP mutational spectrum is variegated and characterized by point mutations (30-50 %) and deletions (~10%). The latter are diverse in size and genomic position and remove either the whole CREBBP gene and its flanking regions or only an intragenic portion. Here, we report 14 novel CREBBP deletions ranging from single exons to the whole gene and flanking regions which were identified by applying complementary cytomolecular techniques: fluorescence in situ hybridization, multiplex ligation-dependent probe amplification and array comparative genome hybridization, to a large cohort of RSTS patients. Deletions involving CREBBP account for 23% of our detected CREBBP mutations, making an important contribution to the mutational spectrum. Genotype-phenotype correlations revealed that patients with CREBBP deletions extending beyond this gene did not always have a more severe phenotype than patients harboring CREBBP point mutations, suggesting that neighboring genes play only a limited role in the etiopathogenesis of CREBBP-centerd contiguous gene syndrome. Accordingly, the extent of the deletion is not predictive of the severity of the clinical phenotype.

Insights into genotype-phenotype correlations from CREBBP point mutation screening in a cohort of 46 Rubinstein-Taybi syndrome patients

The genetic basis of Rubinstein-Taybi syndrome (RSTS), a rare, sporadic, clinically heterogeneous disorder characterized by cognitive impairment and a wide spectrum of multiple congenital anomalies, is primarily due to private mutations in CREBBP (approximately 55% of cases) or EP300 (approximately 8% of cases). Herein, we report the clinical and the genetic data taken from a cohort of 46 RSTS patients, all carriers of CREBBP point mutations. Molecular analysis revealed 45 different gene alterations including 31 inactivating (21 frameshift and 10 nonsense), 10 missense and 4 splicing mutations. Bioinformatic tools and transcript analyses were used to predict the functional effects of missense and splicing alterations. Of the 45 mutations, 42 are unreported and 3 were described previously. Recurrent mutations maybe a key tool in addressing genotype-phenotype correlations in patients sharing the same defects (at the genomic or transcript level) and specific clinical signs, demonstrated here in two cases. The clinical data of our cohort evidenced frequent signs such as arched eyebrows, epicanthus, synophrys and/or frontal hypertrichosis and broad phalanges that, previously overlooked in RSTS diagnosis, now could be considered. Some suggested correlations between organ-specific anomalies and affected CREB-binding protein domains broaden the RSTS clinical spectrum and perhaps will enhance patient follow-up and clinical care.

Clinical and molecular characterization of Rubinstein-Taybi syndrome patients carrying distinct novel mutations of the EP300 gene

Rubinstein-Taybi syndrome (RSTS) is a rare congenital neurodevelopmental disorder characterized by postnatal growth deficiency, skeletal abnormalities, dysmorphic features and cognitive deficit. Mutations in two genes, CREBBP and EP300, encoding two homologous transcriptional co-activators, have been identified in ˜55% and ˜3-5% of affected individuals, respectively. To date, only eight EP300-mutated RSTS patients have been described and 12 additional mutations are reported in the database LOVD. In this study, EP300 analysis was performed on 33 CREBBP-negative RSTS patients leading to the identification of six unreported germline EP300 alterations comprising one deletion and five point mutations. All six patients showed a convincing, albeit mild, RSTS phenotype with minor skeletal anomalies, slight cognitive impairment and few major malformations. Beyond the expansion of the RSTS-EP300-mutated cohort, this study indicates that EP300-related RSTS cases occur more frequently than previously thought (˜8% vs 3-5%); furthermore, the characterization of novel EP300 mutations in RSTS patients will enhance the clinical practice and genotype-phenotype correlations.