Evolutionary computing in recommender systems: a review of recent research

One of the main current applications of intelligent systems is recommender systems (RS). RS can help users to find relevant items in huge information spaces in a personalized way. Several techniques have been investigated for the development of RS. One of them is evolutionary computational (EC) techniques, which is an emerging trend with various application areas. The increasing interest in using EC for web personalization, information retrieval and RS fostered the publication of survey papers on the subject. However, these surveys have analyzed only a small number of publications, around ten. This study provides a comprehensive review of more than 65 research publications focusing on five aspects we consider relevant for such: the recommendation technique used, the datasets and the evaluation methods adopted in their experimental parts, the baselines employed in the experimental comparison of proposed approaches and the reproducibility of the reported experiments. At the end of this review, we discuss negative and positive aspects of these papers, as well as point out opportunities, challenges and possible future research directions. To the best of our knowledge, this review is the most comprehensive review of various approaches using EC in RS. Thus, we believe this review will be a relevant material for researchers interested in EC and RS.     

Portugal's contribution to navigation and discovery of the New World

Even before the birth of Prince Henry the Navigator (1394) Portugal had displayed a maritime calling due to its 500-mile shore line and numerous natural bays. Inspired by the riches of India he saw during the Portuguese exploration along the coast of West Africa, Prince Henry set out methodically to collect information by bringing together Jews and Moors with geographical knowledge to found his School of Navigation. From 1415 until his death in 1460 he attracted to his school the foremost contemporary scholars in mathematics, astronomy, and cartography along with experts in knowledge of the compass, astrolabe, water currents, and the winds. This concentration of talent yielded the invention of the caravel, the most important navigational advancement of the time, crucial for long voyages across the high seas. Although the rest of Europe busied itself in political and religious wars, for 70 years (1415-1492) Portugal alone pursued the discovery of the Atlantic. "No nation in the 15th century exhibited so great a spirit of maritime enterprise as the Portuguese."     

Treatment of acute lymphoblastic leukemia. 30 years' experience at St. Jude Children's Research Hospital

BACKGROUND: Therapy for childhood lymphoblastic leukemia has evolved during the past three decades, but key questions about what are the least toxic, most effective forms of treatment remain unanswered because of the lack of comprehensive follow-up information. METHODS: To assess long-term outcome in the series of clinical trials conducted at St. Jude Hospital, we compared the results of treatment typical of four eras: exploratory combination chemotherapy (era 1, 1962 to 1966; 91 patients), regimens for the control of meningeal leukemia (era 2, 1967 to 1979; 825 patients), limited intensification of therapy (era 3, 1979 to 1983; 428 patients), and extended intensification of therapy (era 4, 1984 to 1988; 358 patients). ("Intensification" refers to strategies of systemic chemotherapy that are more aggressive than conventional ones.) The major end points were survival and event-free survival; we also calculated the relative risk of treatment failure and the rate of relapse or death after treatment ended (post-treatment failure rate). RESULTS: The probability of event-free survival improved significantly in each successive era (P < 0.001 by the log-rank test), reaching 71 percent in era 4. There was a decrease of approximately 50 percent in the risk of treatment failure from one era to the next in each subgroup of patients defined according to different combinations of the leukocyte count, race, age, and sex. Leukemia appeared to be eradicated in patients who remained in complete remission for three years or more after treatment in era 4. The incidence of death due to nonleukemic causes remained 4 to 6 percent despite the trend toward more intensive treatment. An estimated 765 patients (45 percent) are long-term survivors; most of them (80 percent) have no health problems related to leukemia or its treatment. CONCLUSIONS: The development and successful application of preventive therapy for meningeal leukemia, followed by the intensification of systemic chemotherapy, has progressively improved the rate of cure of childhood lymphoblastic leukemia, with relatively few adverse sequelae.     

Reliability of seizure classification using a semistructured interview

Methods for standardized classification of epileptic seizures are important for both clinical practice and epidemiologic research. In this study, we developed a strategy for standardized classification using a semistructured telephone interview and operational diagnostic criteria. We interviewed 1,957 adults with epilepsy ascertained from voluntary organizations. To confirm and expand the seizure history, we also interviewed a first-degree relative for 67% of subjects and obtained medical records for 59%. Three lay reviewers used all available information to classify seizures. To assess reliability, each reviewer classified a sample of subjects assigned to the others. In addition, an expert physician classified a sample of subjects assigned to two of the reviewers. Agreement was "moderate-substantial" for generalized-onset seizures, both for the comparisons between pairs of lay reviewers and for the neurologist versus lay reviewers. Agreement was "substantial-almost perfect" for partial-onset seizures, both for pairs of lay reviewers and for the neurologist versus lay reviewers. These results suggest that seizures can be reliably classified by lay reviewers, using operational criteria applied to symptoms ascertained in a semistructured telephone interview.     

Raman microprobe analysis of preforms and optical fibers

The concentration profile of various dopants (germanium, phosphorus, and fluorine) in preforms and optical fibers has been obtained with a Raman microprobe. A 2-microm spatial resolution was achieved. In the case of germanium and phosphorus, the results agree quite well with those obtained with an electron microprobe. Raman spectroscopy easily detects fluorine. From measurements of various F-doped samples, diffusion of fluorine in undoped and doped silica is suggested.     

Kidney injury and cell therapy: Preclinical study

The aim of this study is to show histological and immunofluorescence analysis of renal parenchyma of agoutis affected by gentamicin‐induced renal disease after the infusion of bone marrow mononuclear cells (BMMC) stained with Hoechst®. Nine agouti's males were divided into three groups: Test group (TG): renal disease by gentamicin induced (n = 3), cell therapy group (CTG): renal disease by gentamicin induced and BMMC infusion (n = 3), and control group (CG): nonrenal disease and BMMC infusion (n = 3). TG and CTG were submitted to the protocol of renal disease induction using weekly application of gentamicin sulfate for 4 months. CG and CTG received a 1 × 108 BMMC stained with Hoechst and were euthanized for kidney examination 21 days after BMMC injection and samples were collected for histology and immunofluorescence analysis. Histological analysis demonstrated typical interstitial lesions in kidney similarly to human disease, as tubular necrosis, glomerular destruction, atrophy tubular, fibrotic areas, and collagen deposition. We conclude that histological analysis suggest a positive application of agouti's as a model for a gentamicin inducing of kidney disease, beyond the immunofluorescence analysis suggest a significant migration of BMMC to sites of renal injury in CTG. Microsc. Res. Tech., 2012. © 2011 Wiley Periodicals, Inc.     

Enhanced default risk models with SVM+

Default risk models have lately raised a great interest due to the recent world economic crisis. In spite of many advanced techniques that have extensively been proposed, no comprehensive method incorporating a holistic perspective has hitherto been considered. Thus, the existing models for bankruptcy prediction lack the whole coverage of contextual knowledge which may prevent the decision makers such as investors and financial analysts to take the right decisions. Recently, SVM+ provides a formal way to incorporate additional information (not only training data) onto the learning models improving generalization. In financial settings examples of such non-financial (though relevant) information are marketing reports, competitors landscape, economic environment, customers screening, industry trends, etc. By exploiting additional information able to improve classical inductive learning we propose a prediction model where data is naturally separated into several structured groups clustered by the size and annual turnover of the firms. Experimental results in the setting of a heterogeneous data set of French companies demonstrated that the proposed default risk model showed better predictability performance than the baseline SVM and multi-task learning with SVM.     

Preference-based multi-objective evolutionary algorithms for power-aware application mapping on NoC platforms

Network-on-chip (NoC) are considered the next generation of communication infrastructure in embedded systems. In the platform-based design methodology, an application is implemented by a set of collaborative intellectual property (IP) blocks. The selection of the most suited set of IPs as well as their physical mapping onto the NoC infrastructure to implement efficiently the application at hand are two hard combinatorial problems that occur during the synthesis process of Noc-based embedded system implementation. In this paper, we propose an innovative preference-based multi-objective evolutionary methodology to perform the assignment and mapping stages. We use one of the well-known and efficient multi-objective evolutionary algorithms NSGA-II and microGA as a kernel. The optimization processes of assignment and mapping are both driven by the minimization of the required silicon area and imposed execution time of the application, considering that the decision maker’s preference is a pre-specified value of the overall power consumption of the implementation.     

Computer simulation of the precordial QRS complex: effects of simulated changes in ventricular wall thickness and volume

Over a 4-year period 40,923 operations and 44,716 surgical admissions were monitored for both community and hospital onset infections. One thousand eight hundred sixty-five patients had 1966 surgical wound infections and 2056 remote infections including 1652 hospital onset and 404 community onset infections. One thousand one hudnred forty-four patients with multiple infections averaged 40 days in the hospital contrasted with 24 days for 721 patients with a single wound infection. The total excess cost of hospitalization for these patients was $951,150. A statistically significant reduction occurred for urinary tract infections, lower respiratory infections and clean and contaminated surgical wound infections. It is suggested that these are all inter-related and a significant reduction in surgical wound infections can be achieved through control of infections at remote sites, particularly those associated with medical devices. The coagulase positive staphylococcus is still the most important single bacterial species in the primary etiology of surgical wound infections. When the gastrointestinal tract is entered or "supra" infecting organisms appear, gram negative bacteria and mixed gram negative and gram positive infections are dominant. Reduction in remote site infections occurring in surgical patients is necessary to reduce the incidence of surgical wound infections, suggest preventive and control measures, and document the effectiveness of such measures.