Fundamentals
Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam in Introduction to Computational Health Informatics, 2019
The important thing is the how tight and effective this binding is. This requires computational model to discover the best configuration of the antibody that would have strong binding. This is done using two techniques: 1) 3D computational model and 2) affinity analysis. Three-dimensional modeling approximates the three-dimensional folding structure of proteins using simulation. Affinity analysis derives the strength of binding of amino-acids under a specific protein configuration.
Which set of factors contribute to increase the likelihood of pedestrian fatality in road crashes?
Published in International Journal of Injury Control and Safety Promotion, 2018
Mohammad Mehdi Besharati, Ali Tavakoli Kashani
The discovery of association rules is one of the major techniques of descriptive data mining, and it is perhaps the most common form of local-pattern discovery in unsupervised learning systems (Kantardzic, 2011). This technique has been frequently used to uncover hidden patterns in a variety of fields, especially in market basket analysis. Traffic safety researchers have also benefitted the association rules technique to detect the associations and interdependencies among the contributory factors of traffic crashes (Geurts, Wets, Brijs, & Vanhoof, 2003; Montella, 2011; Weng, Zhu, Yan, & Liu, 2016). For example, Montella (2011) employed an association discovery technique to explore the interdependencies between different crash types in urban roundabouts and the contributory factors. As stated in Pande and Abdel-Aty (2009), association rules mining is an efficient tool for analysing huge crash database from jurisdictions such as a countrywide crash database.
Critical appraisal of jointness concepts in Bayesian model averaging: evidence from life sciences, sociology, and other scientific fields
Published in Journal of Applied Statistics, 2018
The jointness measures validated in the present study have been constructed for the context of multiple regression with a response and a (potentially large) number of predictors. In this setting, the goal is to assess interrelations between predictors. In contrast to this research question, concepts from the field of association rule analysis in data mining focus on correlation in the context of market basket analysis. For example, one question posed could be whether a consumer buying milk and eggs jointly is likely to buy bread also, see Geng and Hamilton [17]. Or, by analyzing market basket data, a finding could be that a given percentage of transactions that include coffee also include milk, see Wu et al. [45]. Given this background, binary data features prominently in association rule analysis, see, for example, Plasse et al. [30], while regression problems considered in jointness applications typically involve a mix of all data types (continuous, categorical, binary).
Analysis of contributory factors of fatal pedestrian crashes by mixed logit model and association rules
Published in International Journal of Injury Control and Safety Promotion, 2023
Maria Rella Riccardi, Filomena Mauriello, Antonella Scarano, Alfonso Montella
The association rules can generally be seen as a sort of market basket analysis, where the contents of each shopper’s cart are recorded to analyse the co-occurrence of purchased items (Pande & Abdel-Aty, 2009). The main feature of association rules is that they allow the generation of easily understandable patterns discovering interesting relationships between various attributes in the crash dataset. Using the a priori algorithm (Agrawal et al., 1993), simple steps are repeated to examine available data and identify frequently occurring item-sets, sub-sequences, or arrangements from large database (Das et al., 2019; Xu et al., 2018).
Related Knowledge Centers
- Cluster Analysis
- Data Mining
- Association Rule Learning
- Evidence-Based Medicine