THE RESEARCH OF PREDICTING THE CERVICAL CANCER BASING ON BAYESIAN ALGORITHM
Journal Title: Topics in Intelligent Computing and Industry Design (ICID) - Year 2017, Vol 1, Issue 3
Abstract
Cervical cancer remains a significant cause of mortality in low-income country. It is necessary to find a most effective machine learning algorithm to predict the cervical cancer. Our research attempts to help doctor to judge whether a patient develops cervical cancer. Thus, we use Bayesian Classification, decision tree and Support Vector Machine to analysis the dataset. As a result, we compared 0-1 loss classification error rate of these algorithm, whose error rate is the least so it is the best. And the conclusion is that AODE is the most suitable algorithm of these.
Authors and Affiliations
Bei Hui, YuNan Jiang, HongTing Zhou, Lin Ji, Jia Chen,
Keywords
Related Articles
- EP ID EP409145
- DOI 10.26480/iscsai.01.2017.20.22
- Views 73
- Downloads 0
ALGORITHM RESEARCH ON CONFLICT DETECTION AND ELIMINATION IN COLLABORATIVE WORK
For cross-enterprise resource conflict problem in collaborative work plan, put forward a new kind of conflict detection and elimination of cross-enterprise project work model, based on the time of supply and demand plan...
STUDY ON THE PRINCIPLE AND APPLICATION OF PIEZOELECTRIC CERAMICS ENERGY SUPPLY TECHNOLOGY
Piezoelectric ceramic power generation technology to meet the requirements of environmental self-adaptive power supply. It has a long life, pollution-free, maintenance-free advantages, a broad application prospects. This...
A NEW METHOD OF CORRELATION REGION LEAST SQUARE ALGORITHM FOR EARTH SYSTEM PARAMETER IDENTIFICATION
In the earth exploration field, least square algorithm is one of the most commonly traditional methods for earth parameter system identification. Aiming at the problems of the traditional least square algorithm based on...
ANALYSING FACTORS INFLUENCING E-GOVERNMENT DEVELOPMENT IN ZAMBIA: A PRINCIPAL COMPONENT ANALYSIS APPROACH
Effervescent e-Government development entails that e-Government applications and solutions are accessed by a majority of citizens and businesses accessing many of government information or services and participating in t...
VEHICLE DATA PROCESSING AND ANALYSIS PLATFORM BASED ON SPARK
Based on the Spark big data analysis platform, using Yarn management resource scheduling problem, using Hbase as the storage mode of distributed data. Through the mass of car data, dig out the key factors that affect the...