China 
Africa 
Explore chapters and articles related to this topic
Big Data Analytics for COVID-19
Published in Gitanjali Rahul Shinde, Asmita Balasaheb Kalamkar, Parikshit N. Mahalle, Nilanjan Dey, Data Analytics for Pandemics, 2020
Gitanjali Rahul Shinde, Asmita Balasaheb Kalamkar, Parikshit N. Mahalle, Nilanjan Dey
Discussion: Considering the example stated above, the association rule learning technique is used when the dataset consists of logical interrelations between multiple columns or features. This technique is useful to explore the correlation among various columns of a database or dataset. This technique uses a rule-based (if/then or if/then/else) mechanism to infer the associations or logical connections between various columns or features in the dataset. Support and confidence are the two main patterns majorly used by this technique in order to find semantic similarities. Apriori, Eclat, and F-P growth algorithms [29] are popular algorithms used for association rule learning.
Intelligent Data Analysis Techniques
Published in Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam, Introduction to Computational Health Informatics, 2019
Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam
Associative learning derives associative-rules of the form antecedent → consequent. It is useful in data analytics for the discovery of new and sometimes unusual patterns. Apriori algorithm is a popular algorithm to derive association rules. The formation of a rule is based upon forming the associative set of variables and splitting the associative set into two subsets: antecedent and consequent. Only those variables are selected that have support above a minimum threshold. Similarly, only those associated rules are selected that have confidence above a threshold. Multiple rules are merged if they share the same prefix. Associated rules may exhibit false correlation, if the support is high. To handle false correlations, lift-ratio is used that is a ratio of support of the association set and the product of the support of antecedent and consequent. A lift-ratio > 1.0 shows false correlation. Association rules are used in identifying the association of disease, the lab-tests and biomarkers.
Swarm Intelligence and Evolutionary Algorithms for Heart Disease Diagnosis
Published in Sandeep Kumar, Anand Nayyar, Anand Paul, Swarm Intelligence and Evolutionary Algorithms in Healthcare and Drug Development, 2019
Association Rules: The objective in association rule-based mechanism is to obtain a hidden pattern between two different elements during each instance of their transactions. In case of predicting heart disease, the association rule is used to explain, analyze and classify the various parameters recorded for each patient. Then forecast the heart disease based on the risk factor derived using various parameters [31].
Target association rule mining to explore novel paediatric illness patterns in emergency settings
Published in Scandinavian Journal of Clinical and Laboratory Investigation, 2022
Pradeep Kumar Dabla, Kamal Upreti, Divakar Singh, Anju Singh, Jitender Sharma, Aashima Dabas, Damien Gruson, Bernard Gouget, Sergio Bernardini, Evgenija Homsak, Sanja Stankovic
Association Rule Mining based system may accept a real number and category data as input and generate rules for a specific class value; that is, clinical features with actual numbers can be utilised directly to find meaningful patterns. The market basket analysis-based targeted association rule mining concept can prove to be useful analysing patient behaviour regarding treatment, patient disease and clinical features, and medicine or therapy success patterns. The rules have different severe responsibility for chronic conditions and the outcome of survival and death of patients demonstrate the need of detecting diverse signs. In the principles discovered for patient discharge and death, we used association rule mining to differentiate consecutive boundaries and define designs. The following common clinical parameters were observed, with their specific value ranges liable for deaths: ventilated, ventilation duration, GCS, platelet count, temperature, diagnosis, lactate levels, TCO2, HCO3, WBC, pCO2, gas K+, gas Na+, systolic blood pressure (SBP), calcium(i), glucose, Mg(i), heart rate, BUN, creatinine, prothrombin time (PT). This study will be helpful in contributing to the development of unique and practical rules for speeding up further research and helping clinicians for better monitoring and decision making of sepsis or critically ill patients.
The development and validation of a human factors analysis and classification system for the construction industry
Published in International Journal of Occupational Safety and Ergonomics, 2022
The Apriori algorithm aims at determining the association rules between two variables. Association rules mean that the occurrence of an event in one set implies the occurrence of events in another set [41]. The Apriori algorithm was originally invented for the problem of how to discover items purchased together with a sufficiently high frequency from large data. Recently, some papers have adopted it for establishing correlations. Yang and Chen [42] employed the Apriori algorithm as one of the resources for finding the correlation between clinical information and pathology reports to support lung cancer pathologic staging diagnoses. Wu and Li [43] developed a dynamic mining algorithm using the Apriori algorithm to quantify the alarm correlation in communication networks. As a result, it is promising to utilize the Apriori algorithm to determine the correlations among variables.
Discovering opioid users’ medical comorbidities: a data mining approach
Published in Journal of Substance Use, 2020
Yong-Mi Kim, Pranay Kathuria, Dursun Delen
This analytical strategy could discover millions of association rules due to the number of medical comorbidities in the dataset. The rapidly growing number of rules can be pruned (filtered) using support and confidence levels. The support of a rule is the proportion of transactions (i.e., total medical records) that contain both X (opioids) and Y (high BP, high glucose) in the itemset. It is expressed as supp(X→Y) = σ(X∪Y)/N. N is the total number of the medical records in the analysis. It is used to discover how frequently these items occur together out of the total number of medical records.