Study on the Application of Data Mining in the Reliability Evaluation of Structures

2011 ◽  
Vol 243-249 ◽  
pp. 5699-5702
Author(s):  
Wei Bing Hu ◽  
Yan Fang Hou
Keyword(s):  
Data Mining ◽  
Reliability Evaluation ◽  
Fuzzy Cluster ◽  
New Method ◽  
Factors Affecting ◽  
New Information ◽  
Input Parameters ◽  
Structure Code

This paper studies several factors that affect the reliability of established structures .The theory of Data Mining is introduced into this paper as a new method to study the evaluation of established structures. New information is ultimately mined from “Data Sea” filed up in existed structures. According to the Chinese Structure Code and engineering experience, the factors affecting the reliability of structures are properly qualified as input parameters. This paper mainly researches with engineering structural examples on the theory of fuzzy cluster, which belong to the field of Data Mining. According to the analysis of the methods， it is found that the method put forward in this paper is practicable, reliable and effective.

The Application of Neural Networks in the Reliability Evaluation of Structures

2010 ◽  
Vol 452-453 ◽  
pp. 497-500
Author(s):  
Wei Bing Hu ◽  
Hai Pin Yu
Keyword(s):  
Neural Networks ◽  
Reliability Evaluation ◽  
New Method ◽  
Traditional Methods ◽  
Factors Affecting ◽  
Long Period ◽  
Input Parameters ◽  
Structure Code

According to the Chinese Structure Code and engineering experience, the factors affecting the reliability of structures are properly qualified as input parameters; this paper mainly researches the application of neural networks in the reliability evaluation of structures. Comparisons are made between the traditional methods of reliability evaluation of established structures and the method put forward in this paper. It is found that the method put forward in this paper is practicable, reliable and effective. Meanwhile, the weaknesses of the traditional reliability evaluation of structures, that is, long period of evaluation, huge and tedious calculation and arbitral judgments, are avoided with the introduction of the new method in this paper.

scholarly journals Applying Improved Apriori Algorithm in Figuring out the Relation between Weather Factors and Rainfall

2020 ◽  
Vol 1 (1) ◽  
pp. 23-26
Author(s):  
Siti Zulaikha ◽  
Martaleli Bettiza ◽  
Nola Ritha
Keyword(s):  
Data Mining ◽  
Big Data ◽  
Light Intensity ◽  
Apriori Algorithm ◽  
Weather Factors ◽  
Factors Affecting ◽  
Data Store ◽  
Repeated Pattern ◽  
Major Factors ◽  
Sun Light

Data on the rainfall is compelling to study as it becomes one of the major factors affecting the weather in a certain region and various aspects of life as well. Generally, predicting rainfall is performed by analyzing data in the past in certain methods. Rainfall is prone to follow repeated pattern in sequence of time. The utilization of big data mining is expected to result in any valuable information that used to be unrevealed in the big data store. Some methods used in data mining are Apriori Algorithm and Improved Apriori Algorithm. Improved Apriori itself is to represent the database in the form of matrix to describe its relation in the database. Data used in this research is the rainfall factor in 2016 in Tanjungpinang city. Based on the test of Improved Apriori Algorithm, it was found out that the relation of the rainfall and weather factors utilizing 2 item sets, that is, if the temperature is low (24,0 - 26,0), the humidity is high (85 - 100), then the rainfall is mild. If the temperature is low (24,0 - 26,0), the light intensity is low (0 – 3), then the rainfall is heavy, and 3 item sets if the temperature is low (24,0 - 26,0), the humidity is high (85 - 100), the sun light intensity is low (0-3), then the rainfall is medium.

scholarly journals Hybrid Method for Detecting Duplicate Image by Using Image Retrieval Technique in Data Mining

2021 ◽  
Vol 09 (04) ◽  
Author(s):  
Dr. S. Thavamani ◽  
Keyword(s):  
Data Mining ◽  
Image Retrieval ◽  
Hybrid Method ◽  
Video Retrieval ◽  
New Method ◽  
Computational Time ◽  
Storage Space ◽  
Copy Detection ◽  
Retrieval Technique ◽  
Improved Accuracy

Duplicated images cause several problems in online sites, so these demand special attention. To address the disadvantages of frames copy detection, the Hybrid Method of Detecting Duplicate Image by Using Image Retrieval Technique in Data Mining was proposed. We use the new method of eliminating duplicates in this example. To address the disadvantages of frames copy detection, the Hybrid Method of Detecting Duplicate Image by Using Image Retrieval Technique in Data Mining was proposed. The new method of eliminating duplicates in this example has proposed. Using this method, you can get rid of frames that aren't relevant to the video. This makes for more precise and faster video retrieval with fewer duplicates. As a back end, this technique is implemented in C# and SQL. The findings are put to the test and compared to the current SIFT process. The results showed that the output improved accuracy while reducing storage space, computational time, and memory use.

scholarly journals Factors Affecting Birthweight Decline in Recent Japan Based on Birth Certificates

2020 ◽  
Author(s):  
Noriko Kato ◽  
Catherine Sauvaget ◽  
Honami Yoshida ◽  
Tetsuji Yokoyama ◽  
Nobuo Yoshiike
Keyword(s):  
Birth Order ◽  
Gestational Age ◽  
Maternal Diet ◽  
Linear Regression Analysis ◽  
Vital Statistics ◽  
Birth Certificates ◽  
Factors Affecting ◽  
New Information ◽  
Multivariable Linear Regression Analysis ◽  
Birth Records

Abstract Background:Birthweight is declining consistently for more than 30 years in Japan. Rapid rise in low birth weight infant counts one of the worst among OECD countries.Objective: To add new information for clarifying the factors associated with the decline in birthweight in Japan.Methods: Government vital statistics records were used under permission. 40,968,266 birth records born between 1980 and 2004 were analyzed. Multivariable linear regression analysis was used to examine whether the decline in the birthweight could be explained by obstetrical variables such as gestational age and plurality.Results: From 1980 to 2004, we observed a decline in mean birthweight with yearly effect of -8.07g, which got steeper after 1985 and persisted until 1999, and plateaued thereafter. After adjustment for gestational age, neonatal gender, birth order, plurality, father age, yearly effect became -5.13g, between 1980 and 2004. Conclusion：Recent decreases in birthweight among Japanese neonates were not fully explained by trends of gestational age, sex, birth order, plurality and father age. We should consider additional factors such as pre-pregnant maternal BMI and maternal diet.

scholarly journals FACTORS AFFECTING THE CUSTOMS CLEARANCE TIME AT PRIME CUSTOMS OFFICE TYPE A OF TANJUNG PRIOK

2020 ◽  
Vol 2 (2) ◽  
pp. 01-17
Author(s):  
Khamami Herusantoso ◽  
Ardyanto Dwi Saputra
Keyword(s):  
Data Mining ◽  
Type A ◽  
Database Analysis ◽  
Data Set ◽  
Complex Phase ◽  
Specific Data ◽  
Factors Affecting ◽  
Efficiency And Effectiveness ◽  
Hidden Knowledge ◽  
Clearance Process

In the dwell-time, the customs clearance is considered as the most complex phase, even though its portion is the shortest among other phases, such as pre-clearance and post clearance. In order to improve the efficiency and effectiveness on the services performed in the customs clearance process, the customs authorities must start considering the help of database analysis in identifying obstacles instead of depending on the personal analysis. Useful information is hidden among the importation data set and it is extractable through data mining techniques. This study explores the customs clearance process of import cargo whose document is declared through the red channel at Prime Customs Office Type A of Tanjung Priok (PCO Tanjung Priok), and applies a specific data mining classifier called the decision tree with J48 algorithm to evaluate the process. There are 11 classification models developed using unpruned, online pruning, and post-pruning features. One best model is chosen to extract the hidden knowledge that describes factors affecting the customs clearance process and allows the customs authorities to improve their services performed in the future.

AN INFORMATION-THEORETIC FILTER METHOD FOR FEATURE WEIGHTING IN NAIVE BAYES

2014 ◽  
Vol 28 (05) ◽  
pp. 1451007 ◽  
Author(s):  
CHANG-HWAN LEE
Keyword(s):  
Data Mining ◽  
Bayesian Learning ◽  
State Of The Art ◽  
Feature Weighting ◽  
New Method ◽  
Filter Method ◽  
Information Theoretic ◽  
Naive Bayesian ◽  
Naïve Bayesian ◽  
Unrealistic Assumption

In spite of its simplicity, naive Bayesian learning has been widely used in many data mining applications. However, the unrealistic assumption that all features are equally important negatively impacts the performance of naive Bayesian learning. In this paper, we propose a new method that uses a Kullback–Leibler measure to calculate the weights of the features analyzed in naive Bayesian learning. Its performance is compared to that of other state-of-the-art methods over a number of datasets.

Defining Knowledge Constituents and Contents

2016 ◽  
Vol 5 (1) ◽  
pp. 1-7
Author(s):  
Sead Spuzic ◽  
Ramadas Narayanan ◽  
Megat Aiman Alif ◽  
Nor Aishah M.N.
Keyword(s):  
Artificial Intelligence ◽  
Data Mining ◽  
Information Technology ◽  
Knowledge Extraction ◽  
Information Environment ◽  
Electronic Records ◽  
Software Complexity ◽  
New Information ◽  
Key Terms

While it appears that a consensus is crystalising with regard to the hierarchy of concepts such as “knowledge”, “definition” and “information”, there is an increasing urgency for improving definitions of these terms. Strategies such as “knowledge extraction” or “data mining” rely on the increasing availability of digital (electronic) records addressing almost any aspect of socio-economic realm. Information processors are invaluable in the capacity of turning large amount of data into information. However, a new problem emerged on the surface in this new information environment: numerous concepts and terms are blurred by ambiguous definitions (including the concept of 'definition' itself). This triggered a need for mitigating hindrances such as homonymy and synonymy, leading further to demands on the decoding software complexity of which equals the artificial intelligence applications. Information technology presumably copes with this diversity by providing the information decoding 'tools'. This opens a never-ending opportunity for further permutations of tasks and service abilities. The solution, however, is to address the causes rather than indulge in multiplying the superficial remedies. Clearly, the multiplicity of definitions for the same concepts, false synonyms and so forth show that there is a need for introducing definitions of sufficient dimensionality. In this article, a number of examples of important concepts are presented first to point at the ambiguities associated with them, and then to propose their disambiguation. The minimum intent is to demonstrate how these key terms can be defined to avoid ambiguities such as pleonasm, homonymy, synonymy and circularity.

scholarly journals Research on Factors Affecting Real Estate Values by Data Mining

2018 ◽  
Vol 6 (1) ◽  
pp. 220-239 ◽  
Author(s):  
Filiz Ersoz ◽  
Taner Ersoz ◽  
Muhammet Soydan
Keyword(s):  
Data Mining ◽  
Real Estate ◽  
Statistical Modelling ◽  
Unit Price ◽  
City Centre ◽  
Important Place ◽  
Factors Affecting ◽  
The Real ◽  
The Real Estate ◽  
Real Estate Valuation

Abstract Construction sector has an important place in Turkey’s economy. Real estate sales for the sector are increasing in parallel. However, the purchase cost is also important for those who are willing to buy a real estate. In the acquisition of real estate, factors such as size, location and age of the house are taken into consideration. The aim of the article is to conduct research on factors affecting real estate values by data mining. In this study, the most important variables that determine the value of the real estate have been investigated by data mining methods. The research has been carried out in Karabük and the variables determined according to the opinions of real estate experts. As classification methods, CHAID and C&RT algorithms have been used. It has been evaluated that both algorithm estimation results can be used. Within the framework of the study, the variables that have the most impact on the unit price have been determined, such as the size of the real estate, the distance to the city centre, the popularity, and the age of the building. The use of advanced technologies, such as statistical modelling and machine learning in real estate valuation and automatic value estimation, is of importance in determining the real value of the real estate.

Sign in / Sign up

Export Citation Format

Share Document