China
Africa
Explore chapters and articles related to this topic
Data Science and Machine Learning Applications for Mental Health
Published in Pallavi Vijay Chavan, Parikshit N Mahalle, Ramchandra Mangrulkar, Idongesit Williams, Data Science, 2022
Dhruvi Khankhoje, Pruthav Jhaveri, Narendra M. Shekokar
Today’s generation has seen a significant increase in the number of people suffering from depression. Given their daily life and workload, many people overlook the symptoms of melancholia and depression, this can prove to be very harmful over time and can produce serious complications within individuals. More research is needed to diagnose and treat melancholia and depression in its early stages before it can cause irreversible damage. In this section, a data mining method is discussed, following a questionnaire which was filled by patients data mining processes have been used to extract association rules [23]. The questionnaire aims to understand the physical and mental conditions of the patient by drawing a relationship between how a person feels and acts and the degree of depression he suffers from. The method reviewed advocates the use of Frequent Pattern (FP) tree algorithm over the Apriori algorithm, which reduces execution time; it discovers interesting relations without repeatedly generating candidate itemsets.
The Impact and Usage of Smartphone among Generation Z: A Study Based on Data Mining Techniques
Published in Mohamed Lahby, Utku Kose, Akash Kumar Bhoi, Explainable Artificial Intelligence for Smart Cities, 2021
Mohamed Uwaiz Fathima Rushda, Lakshika S. Nawarathna
Association rule learning is a rule-based ML method for discovering interesting relations between variables in large databases. An association rule is an implication of the form X → Y. The standard definition of association rule is if a transaction includes X then it also has Y, where X and Y are item sets. Many algorithms are generated for association rules. Apriori uses a breadth-first search strategy to count the support of item sets and uses a candidate generation function which exploits the downward closure property of support. It uses a bottom-up approach, where frequent subsets are extended one item at a time a step known as candidate generation, and groups of candidates are tested against the data. Apriori algorithm computes the frequent item sets in several rounds; it is used to predict the rules.
Occupational safety status and countermeasures based on casualty accidents from 2012 to 2018 in Beijing
Published in Chongfu Huang, Zoe Nivolianitou, Risk Analysis Based on Data and Crisis Response Beyond Knowledge, 2019
Fucai Yu, Xuewei Ji, Aizhi Wu, Ming Wen, Yan Liu
Apriori algorithm is a common association analysis algorithm, which can discover frequent item sets meeting special rules. With Apriori algorithm, the correlation of 14 types of factors, including the time, place, industry, type, harmful factor, cause, equipment and qualification certificate, is analyzed. The minimum length of the item sets is 4, the minimum count is 10, and the minimum confidence is 0.8. According to the above law, 547 rules, whose lift is greater than 1, are obtained. As shown in Figure 5, the maximum value of support is 0.142, the maximum value of confidence is 1, and the maximum value of lift is 29.3.
Machine learning approaches for prediction of properties of natural fiber composites : Apriori algorithm
Published in Australian Journal of Mechanical Engineering, 2022
Balasundaram R, Sathiya Devi S, Sakthi Balan G
Support and confidence are the measures to find an association among the items. Hence, support and confidence are varied as indicated in Table 4 for doing experimentation WEKA is used and results are presented. There is no correlation or association is generated for the support of 50 and confidence value of 99. The reason is that, for the 50% of support, the possible number of minimum correlations is 10 (0.5 × 27 = 13.5 ~ 14). Since only 27 datasets alone were considered for this work, there no correlations among the properties for this particular data set, and hence there no rule for 50% of support and 99% confidence. The degree of dominance of various material properties shown in Table 2 was marked as per the values that are segregated under three levels in Table 3. The values indicated in Table 3 were derived from the literature papers. To find complex inter-relations among the properties of materials, machine learning algorithm could be used. Among the available machine learning techniques, Apriori algorithm is chosen for mining frequent itemsets and to formulate association rules from a large database. Till now there are very limited research works were done on material selection criteria for suitable applications. Only review papers are published in which the output of previously done works were explained. Apriori algorithm is the most simple and easy-to-understand algorithm among association rule learning algorithms. The resulting rules are instinctive and easy to communicate to an end user.
Hit and run crash analysis using association rules mining
Published in Journal of Transportation Safety & Security, 2021
Subasish Das, Xiaoqiang Kong, Ioannis Tsapakis
Market basket analysis is a popular data mining approach. As a nonparametric method, it avoids making any parametric assumptions as most of the parametric methods do. It also has great flexibility while dealing with data sets with a significant amount of variables, which is called frequent itemset (Weng et al., 2016). To tackle the frequent itemset/product problem, researchers invented a large number of algorithms. Among these algorithms, APRIORI, developed by Agrawal and Srikant (1994), is a level-wise, breadth-first algorithm which counts transactions. This algorithm can be used to mine frequent itemsets, maximal frequent itemsets, and closed frequent itemsets. The implementation of the a priori algorithm (principle: if an itemset is frequent, then all of its subsets must also be frequent) can additionally be used to generate association rules.
Association rule learning to improve deficiency inspection in port state control
Published in Maritime Policy & Management, 2020
Wu-Hsun Chung, Sheng-Long Kao, Chun-Min Chang, Chien-Chung Yuan
During the past several decades, various algorithms have been developed for performing association analysis. The Apriori algorithm is a popular algorithm which has been widely used in association analysis. It has been adopted for identifying the association rules between PSC deficiency itemsets in this study. The Apriori algorithm was proposed first by Agrawal and Srikant in (1994). The Apriori algorithm uses a ‘bottom-up’ approach, where frequent subsets are extended one item at a time, and groups of candidates are tested by a given threshold value. The algorithm keeps running until no further qualified extensions are found. The three main parameters used in the algorithm, namely, ‘support’, ‘confidence’, and ‘lift’, are calculated by conditional probability, as shown in Eqs. 1–3, respectively. Figure 2 is an example illustrating how the algorithm works for a transaction database. For the detailed steps of the algorithm, the reader may refer to Agrawal and Srikant (1994).