Genes, environment, and dyslexia the 2005 Norman Geschwind memorial lecture     

Olson RK 《Annals of dyslexia》2006,56(2):205-238

This article presents an overview of some methods and results from our continuing studies of genetic and environmental influences on dyslexia, and on individual differences across the normal range that have been conducted over the past 25 years in the Colorado Learning Disabilities Research Center (CLDRC) and in related projects. CLDRC investigators compare the similarities of identical twin pairs who share all their genes and fraternal twins who share half their segregating genes to assess the balance of genetic, shared family environment, and nonshared environment influences on dyslexia and on individual differences across the normal range. We have learned that among the children we have studied in Colorado, group deficits in reading (dyslexia) and individual differences in reading across the normal range are primarily due to genetic influences, and these genetic influences are often shared with some of the same genetic influences on deficits and individual differences in language and ADHD. We have also learned from our molecular-genetic linkage studies that there are regions on several chromosomes likely to contain genes that influence dyslexia. Several specific genes within these regions have been tentatively identified through molecular-genetic association analyses, but much more research is needed to understand the pathways among specific genes, regions of noncoding DNA that regulate the activity of those genes, the brain, and dyslexia. I conclude with a discussion of our research on individual differences in early reading development, on the role of early learning constraints in dyslexia, and on how genetic influences are expressed through their interaction and correlation with the environment.  相似文献   

Retention of EBM competencies     

Dorsch J  Aiyer MK  Gumidyala K  Meyer LE 《Medical reference services quarterly》2006,25(3):45-57

This longitudinal follow-up study surveyed the attitudes toward, and skills in, evidence-based medicine (EBM) of medical school graduates who had participated in a formal EBM curriculum during their third-year Internal Medicine (IM) clerkship. The objective was to measure whether competencies were retained over time and to identify the long- term impact on the graduates' attitudes toward, and proficiency in, EBM after graduation. Questionnaire results showed that a group of medical school graduates retained EBM skills and a positive attitude about the importance of applying EBM principles in patient care one to three years after completing an EBM course.  相似文献   

Finding related sentence pairs in MEDLINE     

Smith LH  Wilbur WJ 《Information Retrieval》2010,13(6):601-617

We explore the feasibility of automatically identifying sentences in different MEDLINE abstracts that are related in meaning. We compared traditional vector space models with machine learning methods for detecting relatedness, and found that machine learning was superior. The Huber method, a variant of Support Vector Machines which minimizes the modified Huber loss function, achieves 73% precision when the score cutoff is set high enough to identify about one related sentence per abstract on average. We illustrate how an abstract viewed in PubMed might be modified to present the related sentences found in other abstracts by this automatic procedure.  相似文献   

Efficient algorithms for ranking with SVMs   总被引：1，自引：0，他引：1  

O. Chapelle  S. S. Keerthi 《Information Retrieval》2010,13(3):201-215

RankSVM (Herbrich et al. in Advances in large margin classifiers. MIT Press, Cambridge, MA, 2000; Joachims in Proceedings of the ACM conference on knowledge discovery and data mining (KDD), 2002) is a pairwise method for designing ranking models. SVMLight is the only publicly available software for RankSVM. It is slow and, due to incomplete training with it, previous evaluations show RankSVM to have inferior ranking performance. We propose new methods based on primal Newton method to speed up RankSVM training and show that they are 5 orders of magnitude faster than SVMLight. Evaluation on the Letor benchmark datasets after complete training using such methods shows that the performance of RankSVM is excellent.  相似文献   

Statistical query expansion for sentence retrieval and its effects on weak and strong queries     

David E. Losada 《Information Retrieval》2010,13(5):485-506

The retrieval of sentences that are relevant to a given information need is a challenging passage retrieval task. In this context, the well-known vocabulary mismatch problem arises severely because of the fine granularity of the task. Short queries, which are usually the rule rather than the exception, aggravate the problem. Consequently, effective sentence retrieval methods tend to apply some form of query expansion, usually based on pseudo-relevance feedback. Nevertheless, there are no extensive studies comparing different statistical expansion strategies for sentence retrieval. In this work we study thoroughly the effect of distinct statistical expansion methods on sentence retrieval. We start from a set of retrieved documents in which relevant sentences have to be found. In our experiments different term selection strategies are evaluated and we provide empirical evidence to show that expansion before sentence retrieval yields competitive performance. This is particularly novel because expansion for sentence retrieval is often done after sentence retrieval (i.e. expansion terms are mined from a ranked set of sentences) and there are no comparative results available between both types of expansion. Furthermore, this comparison is particularly valuable because there are important implications in time efficiency. We also carefully analyze expansion on weak and strong queries and demonstrate clearly that expanding queries before sentence retrieval is not only more convenient for efficiency purposes, but also more effective when handling poor queries.  相似文献   

Book Distribution in the Arab World     

Jan-C. Eschweiler  Axel Goehler 《Publishing Research Quarterly》2010,26(3):193-201

Business opportunities in international markets are largely determined by accessibility, transparency, and legal frameworks fair to all parties concerned. Arab book markets in principle offer a tremendous potential with more than 300 million readers sharing one common language. Besides unsolved censorship issues, problems with money collection, and differing regulations, however, the lack of a uniform book distribution system is the most critical hurdle for the development of a common Arabic book market. Most importantly, there is no comprehensive database containing accurate content briefs, background information on authors, ISBN numbers, and publisher company information as well as all relevant commercial data such as book prices, weight, logistics terms, and availability. Authors and publishers are hard hit as they often need to organize book transportation, delivery, and invoice collection on their own. At present, “selling book fairs” in the Arab world—with the exception of Abu Dhabi International Book Fair—present themselves as crowded book bazaars rather than executive meeting points. The authors of the article, however, are anticipating change: New distribution initiatives will soon interlink regional Arabic markets and create unique growth and profit opportunities.  相似文献   

Christopher D. Manning,Prabhakar Raghavan,and Hinrich Schütze: Introduction to information retrieval     

I. C. Mogotsi 《Information Retrieval》2010,13(2):192-195

  相似文献   

Entity ranking in Wikipedia: utilising categories,links and topic difficulty prediction     

Jovan Pehcevski  James A. Thom  Anne-Marie Vercoustre  Vladimir Naumovski 《Information Retrieval》2010,13(5):568-600

Entity ranking has recently emerged as a research field that aims at retrieving entities as answers to a query. Unlike entity extraction where the goal is to tag names of entities in documents, entity ranking is primarily focused on returning a ranked list of relevant entity names for the query. Many approaches to entity ranking have been proposed, and most of them were evaluated on the INEX Wikipedia test collection. In this paper, we describe a system we developed for ranking Wikipedia entities in answer to a query. The entity ranking approach implemented in our system utilises the known categories, the link structure of Wikipedia, as well as the link co-occurrences with the entity examples (when provided) to retrieve relevant entities as answers to the query. We also extend our entity ranking approach by utilising the knowledge of predicted classes of topic difficulty. To predict the topic difficulty, we generate a classifier that uses features extracted from an INEX topic definition to classify the topic into an experimentally pre-determined class. This knowledge is then utilised to dynamically set the optimal values for the retrieval parameters of our entity ranking system. Our experiments demonstrate that the use of categories and the link structure of Wikipedia can significantly improve entity ranking effectiveness, and that topic difficulty prediction is a promising approach that could also be exploited to further improve the entity ranking performance.  相似文献   

University of Arkansas for Medical Sciences electronic health record and medical informatics training for undergraduate health professionals     

Jan K Hart  Bruce W Newton  Steven E Boone 《Journal of the Medical Library Association》2010,98(3):212-216

The University of Arkansas for Medical Sciences (UAMS) is planning interprofessional training in electronic health records (EHRs) and medical informatics. Training will be integrated throughout the curricula and will include seminars on broad concepts supplemented with online modules, didactic lectures, and hands-on experiences. Training will prepare future health professionals to use EHRs, evidence-based medicine, medical decision support, and point-of-care tools to reduce errors, improve standards of care, address Health Insurance Portability and Accountability Act requirements and accreditation standards, and promote appropriate documentation to enable data retrieval for clinical research. UAMS will ensure that graduates are ready for the rapidly evolving practice environment created by the HITECH Act.  相似文献   

Networked Services of NIT libraries in India: A study     

Y. Srinivasa Rao  B.K. Choudhury 《Library Collections, Acquisitions, and Technical Services》2010,34(4):105-114

The main purpose of the paper is to find out the provision of providing Networked Services (NSs) among National Institute of Technology (NIT) libraries in India. Methodology used for collecting data was questionnaire. The survey findings indicate that the majority of libraries prefer to provide communication network services (telephone, Internet, e-mail, facsimile etc), online journals (90%), automated catalog system (85%), multimedia databases (85%) and support (70%) services rather than the other services. Concerning the zone wise performance, south zone (80%) libraries are more potential than the other zone libraries in providing NSs.  相似文献