scholarly journals Learning temporal rules to forecast instability in continuously monitored patients

2016 ◽  
Vol 24 (1) ◽  
pp. 47-53 ◽  
Author(s):  
Mathieu Guillame-Bert ◽  
Artur Dubrawski ◽  
Donghan Wang ◽  
Marilyn Hravnak ◽  
Gilles Clermont ◽  
...  
Keyword(s):  
Association Rule ◽  
Continuous Monitoring ◽  
Rule Extraction ◽  
Protein Sequencing ◽  
Face Validity ◽  
Temporal Association ◽  
Use Of Data ◽  
Inductive Machine Learning ◽  
Temporal Association Rule ◽  
Potential Benefits

Inductive machine learning, and in particular extraction of association rules from data, has been successfully used in multiple application domains, such as market basket analysis, disease prognosis, fraud detection, and protein sequencing. The appeal of rule extraction techniques stems from their ability to handle intricate problems yet produce models based on rules that can be comprehended by humans, and are therefore more transparent. Human comprehension is a factor that may improve adoption and use of data-driven decision support systems clinically via face validity. In this work, we explore whether we can reliably and informatively forecast cardiorespiratory instability (CRI) in step-down unit (SDU) patients utilizing data from continuous monitoring of physiologic vital sign (VS) measurements. We use a temporal association rule extraction technique in conjunction with a rule fusion protocol to learn how to forecast CRI in continuously monitored patients. We detail our approach and present and discuss encouraging empirical results obtained using continuous multivariate VS data from the bedside monitors of 297 SDU patients spanning 29 346 hours (3.35 patient-years) of observation. We present example rules that have been learned from data to illustrate potential benefits of comprehensibility of the extracted models, and we analyze the empirical utility of each VS as a potential leading indicator of an impending CRI event.

Fast Algorithms for Temporal Association Rules in a Large Database

2005 ◽  
Vol 277-279 ◽  
pp. 287-292 ◽  
Author(s):  
Lu Na Byon ◽  
Jeong Hye Han
Keyword(s):  
Electronic Commerce ◽  
Association Rules ◽  
Association Rule ◽  
Temporal Association ◽  
Large Database ◽  
Running Time ◽  
Smoothing Filter ◽  
Personalized Services ◽  
Temporal Association Rule ◽  
Transaction Database

As electronic commerce progresses, temporal association rules are developed by time to offer personalized services for customer’s interests. In this article, we propose a temporal association rule and its discovering algorithm with exponential smoothing filter in a large transaction database. Through experimental results, we confirmed that this is more precise and consumes a shorter running time than existing temporal association rules.

Ordering Policy and Inventory Classification Using Temporal Association Rule Mining

2018 ◽  
Vol 6 (1) ◽  
pp. 37-49 ◽  
Author(s):  
Reshu Agarwal
Keyword(s):  
Inventory Management ◽  
Association Rule ◽  
Association Rule Mining ◽  
Temporal Association ◽  
Order Quantity ◽  
Rule Mining ◽  
Ordering Policy ◽  
Temporal Association Rule ◽  
Optimum Order ◽  
Frequent Items

A modified framework that applies temporal association rule mining to inventory management is proposed in this article. The ordering policy of frequent items is determined and inventory is classified based on loss rule. This helps inventory managers to determine optimum order quantity of frequent items together with the most profitable item in each time-span. An example is illustrated to validate the results.

Mining Changes in Temporal Patterns in Latest Time Window for Knowledge Discovery

2019 ◽  
Vol 18 (03) ◽  
pp. 1950028
Author(s):  
Sheel Shalini ◽  
Kanhaiya Lal
Keyword(s):  
Association Rule ◽  
Association Rule Mining ◽  
Time Window ◽  
Search Space ◽  
Temporal Association ◽  
Incremental Mining ◽  
Rule Mining ◽  
Novel Approach ◽  
Temporal Association Rule ◽  
Mining Algorithms

Temporal Association Rule mining uncovers time integrated associations in a transactional database. However, in an environment where database is regularly updated, maintenance of rules is a challenging process. Earlier algorithms suggested for maintaining frequent patterns either suffered from the problem of repeated scanning or the problem of larger storage space. Therefore, this paper proposes an algorithm “Probabilistic Incremental Temporal Association Rule Mining (PITARM)” that uncovers the changed behaviour in an updated database to maintain the rules efficiently. The proposed algorithm defines two support measures to identify itemsets expected to be frequent in the successive segment in advance. It reduces unnecessary scanning of itemsets in the entire database through three-fold verification and avoids generating redundant supersets and power sets from infrequent itemsets. Implementation of pruning technique in incremental mining is a novel approach that makes it better than earlier incremental mining algorithms and consequently reduces search space to a great extent. It scans the entire database only once, thus reducing execution time. Experimental results confirm that it is an enhancement over earlier algorithms.

Sign in / Sign up

Export Citation Format

Share Document