Borrowing Data Mining Based on Association Rules

Using data mining techniques to predict user’s behavior and create recommender systems in the libraries and information centers

“…The results of the study of the loan circulation process of information resources and analysis of library users’ loan transactions using data mining techniques indicate the effectiveness of this new technology in analyzing the mass volume of data in libraries and information centers. The results obtained by using association rules technique, provided a model for users access in the use of information resources and dependency of information resources in different subject categories in terms of using association rules technique to analyze data is similar to the studies (Bussaban and Kularbphettong, 2014; Chen et al , 2008; Huang et al , Jomsri, 2014; Krishnamurthy and Balasubramani, 2014; Long and Wu, 2012; Tsuji et al , 2012; Yu, 2011) and in terms of using the Apriori algorithm is similar to the studies of Krishnamurthy and Balasubramani (2014), Chen et al (2008), Bussaban and Kularbphettong (2014). Moreover, in terms of the selection of rules with high confidence level of above 0.50 are aligned as attractive rules with study of Bussaban and Kularbphettong (2014) and with studies of Chen et al (2008), Huang et al , Long and Wu (2012), which addressed the extraction of the secret dependency rules in the loan circulation records of the libraries.…”

“…Their results also showed that the association rules approach works well in the analysis of large data sets without loss of efficiency (Huang et al ). In another study, using the association rules approach and maximal flock patterns (MFP)-Miner algorithm, Long et al analyzed the transaction data and provided strong rules with the degree of support and confidence (Long and Wu, 2012). In a study to determine the most effective resource recommender systems in libraries, Tsuji et al used three methods based on collaborative filtering, association rules and amazon.…”

“…The results of most studies suggest the effectiveness of the use of various data mining techniques in the field of library sciences and information centers in management affairs and improving the quality of library services. Most studies in this area have used the association rules technique, and their results showed that this technique has a high capability in predicting user transactions and can identify users’ access patterns and predict their transactions (Yu, 2011; Yi et al , 2018; Tsuji et al , 2012; Renaud et al , 2015; Long and Wu, 2012; Liu, 2018; Li and Chen, 2008; Krishnamurthy and Balasubramani, 2014; Jomsri, 2014; Huang et al , Chen et al , 2008; Bussaban and Kularbphettong, 2014). Furthermore, other studies carried out using the neural network and clustering techniques that studied the quality of library services (Li, 2017), users’ satisfaction (Zhang and Wang, 2016) and users’ clustering (Yu, 2011; Kovacevic et al , 2010; Jin and Sheng, 2017; Chen and Chen, 2007; Chen and Chen, 2006) show that these techniques are very appropriate and effective in analyzing the library data.…”

“…Therefore, the use of data mining modern technology in library management systems and information centers is very important (Zhang, 2014; Yi et al , 2014). Academic libraries, as heart of the university and of the main pillars of educational and research activities, use database technologies to facilitate services (Long and Wu, 2012). According to Han, Pei and Kamber, data mining is meant to extract knowledge from huge amounts of data as the most important step in the knowledge discovery whose purpose is exploratory data analysis, discovering patterns, rules and algorithms, predictive modeling and finding deviations (Han et al , 2011).…”

“…According to Han, Pei and Kamber, data mining is meant to extract knowledge from huge amounts of data as the most important step in the knowledge discovery whose purpose is exploratory data analysis, discovering patterns, rules and algorithms, predictive modeling and finding deviations (Han et al , 2011). In another definition, data mining means extracting valuable and potentially useful information contained in the data (Long and Wu, 2012), which, using statistical methods, machine learning and set training, leads to the extraction of accurate information for future management (Bussaban and Kularbphettong, 2014). Some authors, including Chatfield, have expressed data mining as synonyms to the expression “the discovery of knowledge and cognition from the database,” and others, including Fayyad, have expressed it a necessary step in the process of discovering knowledge and cognition from the database (Cox, 1995; Fayyad et al , 1996).…”