Search

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ejor.2005.04.011 Byline: Georgios P. Papamichail (a), Dimitrios P. Papamichail (b) Keywords: Heuristics; Distributed consumer decision-making; Range search; Data clustering; Personalized systems Abstract: This paper describes the k-means range algorithm, a combination of the partitional k-means clustering algorithm with a well known spatial data structure, namely the range tree, which allows fast range searches. It offers a real-time solution for the development of distributed interactive decision aids in e-commerce since it allows the consumer to model his preferences along multiple dimensions, search for product information, and then produce the data clusters of the products retrieved to enhance his purchase decisions. This paper also discusses the implications and advantages of this approach in the development of on-line shopping environments and consumer decision aids in traditional and mobile e-commerce applications. Author Affiliation: (a) Department of Management Science and Technology, Athens University of Economics and Business, 47A Evelpidon & 33 Lefkados Str., Athens 11 362, Greece (b) Computer Science Department, State University of New York at Stony Brook, Stony Brook, NY 11794-4400, USA

Usage of computational methods for design and development of decision support systems, which facilitate group decisions and negotiations in elctronic commerce, is described.

Objective: To investigate a necrotizing enterocolitis (NEC) cluster of late preterm and term neonates (gestational age ≥34 weeks).Methods: We conducted a descriptive and a case-control study. Medical records of neonates with modified Bell stage ≥ IB NEC and matched controls were reviewed, in addition to microbiological and environmental investigation. Study variables included maternal/delivery and neonatal factors, medications, procedures and feeding practices. Univariable/multivariable logistic regression analyses were performed for all and for stage ≥ II cases.Results: Out of 1841 late preterm and term neonates, 10 stage IB and 10 stage ≥ II [mean(SD) birthweight 2529.3 (493.04) g, gestational age 36.96 (1.48) weeks] presented with NEC symptomatology at mean 4.6 (range 2-8) days. Nearly all (19/20) resulted from high-risk pregnancies and received postpartum intermediate care. All were exclusively or partly formula fed. Most (14/20) were born by cesarean delivery. Eight underwent surgery, with no fatality. Intermediate care (p = 0.006), transient tachypnea (p = 0.049), not receiving breast milk (p = 0.019) and in addition intrauterine growth restriction (IUGR) (p = 0.017) for stage ≥ II cases were independently associated with NEC.Conclusions: Late preterm and term neonates in need of intermediate care, with IUGR and transient tachypnea were susceptible to NEC; feeding with breast milk was an important protective factor. [ABSTRACT FROM AUTHOR]

Serotypes and antibiotic sensitivities were determined for 338 strains of Streptococcus pneumoniae from children of north-western Greece with invasive pneumococcal disease (IPD), acute otitis media (AOM) and nasopharyngeal carriage. The most common serotypes among the isolates from IPD were 14 and 19F, while 3, 19F, 9V and 14 were the major cause of AOM. In these groups, the heptavalent conjugate vaccine for pneumococci (7vPCV) seems to cover 90.5% of the serotypes isolated from children less than 2 years old. Serotypes 23F and 6B were the most prevalent in carrier strains. Overall, 23.7% of the isolates were penicillin nonsusceptible (PNS), 97% were fully susceptible to cefotaxime, 29% were resistant to erythromycin, 11.2% to co-trimoxazole and 1.2% to clindamycin. [ABSTRACT FROM AUTHOR]

This paper presents a personalized approach for distributed trust management by employing the k-means range algorithm, a combination of the partitional k-means clustering algorithm with orthogonal range search concepts. The aim of this approach is to aid the human or computer agent in organizing information from multiple sources into clusters according to its "trust features". Thus the agent can perform complicated trust assessments in real-time situations and cooperate with decision-making software to assist in purchasing activities. We conclude by discussing the implications and advantages of this approach in trust management in traditional and mobile e-commerce applications. [ABSTRACT FROM AUTHOR]