[Prognosis for life expectancy using the inductive learning method].

M Rimac; J Kern

PubMed · 10377696

[Prognosis for life expectancy using the inductive learning method].

Abstract

The aim of this paper was to find out the possibility of life expectancy achievement (LEA) prognosis in open population by using epidemiological data, to improve it and to determine the differences between regions. We were using inductive learning software tool, ASSISTANT Professional, based on modified Quinlan's inductive learning method. Data from an epidemiological study of oil/fat consuming influence on diabetes incidence in different regions in Croatia, 1887 examinees, have been used. In spite of limited number of attributes that were available, an improvement in prognosis of LEA has been made by selecting the proper attributes and by changing the limiting values of attributes (values that change the meaning of attributes). We reached the absolute accuracy of 75.54%. Even after further pruning of decision tree that lowered this value, Assistant showed 13% better result than those of random selected outcome. Differences between regions were also established that could not been explained with attributes that were used. Expanding the list of attributes and analysis of their influence in particular region can make further improvement in prognosis of LEA.

Explore related subjects

Croatia Decision Trees Diabetes Mellitus Dietary Fats Epidemiologic Methods Humans Life Expectancy Middle Aged Prognosis Software English Abstract Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

M Rimac, J Kern. [Prognosis for life expectancy using the inductive learning method].. https://pubmed.ncbi.nlm.nih.gov/10377696/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2002

Educatione ad excelentiam--ten years of the Croatian medical journal.

Explore the source record for details and available documents.

Croatia↗

PubMed2002

Scientific biomedical journals in Croatia.

AIM: To assess the current situation in biomedical journal publishing in Croatia, according to the journals editorial and publishing characteristics. METHOD: A survey questionnaire with 38 questions was sent to the editors-in-chief of 43 biomedical journals in Croatia. Thirty-seven journals were analyzed. RESULTS: In 2000, there were 14 general and 23 specialized biomedical journals in Croatia. Twenty-five journals received financial support from the government. Six journals were indexed in MEDLINE, with two of them also in Current Contents, 20 in other bibliographic databases, and 11 were not indexed at all. Twelve journals published articles in English, 11 in Croatian, and 14 in either English or Croatian. Only 9 journals were available on-line: 5 offered abstracts only and 4 offered full-text articles. There were more indexed than non-indexed journals that were financially supported by the government (p=0.031) and published in English or either English or Croatian (p=0.011). Indexed journals published more pages (p=0.020) and received more articles (p=0.030) per year. Five journals provided salary for the editorial staff and 4 paid the reviewers services. Median number of subscribers was 250 (range, 0-7,000) and median circulation 650 (range, 300-7,200). Finances, number and quality of articles, and problems with the review process were perceived by editors as the most important problems. CONCLUSION: Many journals had financial problems despite receiving governmental or other support. Low number/quality of articles received and inefficient review process could reflect the self-containment of Croatian medical community, which comprises around 11,000 physicians and dentists. Large ranges in circulation, number of subscribers, and number of pages published per year, as well as in indexing rates, might imply large differences among the journals in their target readership and editorial policies and performance.

Croatia↗

PubMed2002

Zagreb University School of Medicine: students grades during war.

AIM: To investigate whether examination grades of the students at the Zagreb University School of Medicine changed during the 1991-1995 war in Croatia. METHODS: The retrospective descriptive study included examination grades from 5 major courses in the first 5 years of the medical studies: Anatomy (first year), Physiology (second year), Pathology (third year), Internal Medicine (fourth year), and Surgery (fifth year). We compared the war-period (1991-1995) with two control periods: before (1989-1990) and after the war (1996-2000). The passing grades at the Zagreb University range from 2 to 5, with 2 as the lowest and 5 as the highest grade. There were a total of 17,682 examinations from the 5 courses in the studied periods. RESULTS: Grades were higher in the pre-war and post-war periods than in the war period (p<0.001 for both comparisons). Grades in the individual courses followed the general trend, except for the surgery course, where grades during the war were better than in the control periods, and the physiology course, where continual worsening, irrespective of the war, was observed. CONCLUSION: Medical students obtained significantly lower grades in 5 major courses during the war than before or after it. Thus, it is reasonable to assume that different factors, some related and some unrelated to war, could have adversely influenced students performance.

Croatia↗