V. Janaki

A Novel Approach for Mining Similarity Profiled Temporal Association Patterns Using Venn Diagrams

The problem of mining frequent patterns in a static database is studied extensively in the literature by many researchers. Conventional frequent pattern algorithms are not applicable to find frequent patterns from the temporal database. Temporal database is a database which can store past, present and future information. A temporal relation may be viewed as a database of time invariant and time variant relation instances. The objective of this research is to come up with a novel approach so as to find the temporal association patterns similar to a given reference support sequence and user defined threshold using the concept of Venn diagrams. The proposed approach scans the temporal database only once to find the temporal association patterns and hence reduces the huge overhead incurred when the database is scanned multiple times.

A Survey on Temporal Databases and Data mining

Temporal database is a database which captures and maintains past, present and future data. Conventional databases are not suitable for handling such time varying data. In this context temporal database has gained a significant importance in the field of databases and data mining. The major objective of this research is to perform a detailed survey on temporal databases and the various temporal data mining techniques and explore the various research issues in temporal data mining. We also throw light on the temporal association rules and temporal clustering works carried in literature.

An Approach for Mining Similarity Profiled Temporal Association Patterns Using Gaussian Based Dissimilarity Measure

The problem of mining frequent patterns in non-temporal databases is studied extensively. Conventional frequent pattern algorithms are not applicable to find temporal frequent items from the temporal databases. Given a reference support time sequence, the problem of mining similar temporal association patterns has been the current interest among the researchers working in the area of temporal databases. The main objective of this research is to propose and validate the suitability of Gaussian distribution based dissimilarity measure to find similar and dissimilar temporal association patterns of interest. The measure designed serves as similarity measure for finding the similar temporal association patterns. Finally, in this research, we consider the problem of mining similarity profiled temporal patterns from the set of time stamped transaction of temporal database using proposed measure. We show using a case study how the proposed dissimilarity measure may be used to find the temporal frequent patterns and compare the same with the work carried in the literature. The proposed measure has the fixed lower bound and upper bound values as 0 and 1 respectively which is its advantage as compared to Euclidean distance measure which has no fixed upper bound.

Mining Outlier Temporal Association Patterns

Temporal pattern mining is the recent research among researchers contributing in the areas of data mining, medical mining, spatial data mining, health informatics and gaining significant interest in Internet of things, one of the top ten fields of interest expected to rule the computing world in the next few years. Temporal pattern mining is a knowledge discovery process which concentrates on mining temporal databases for discovering hidden temporal information which includes finding temporal patterns, temporal and spatio- temporal association rules, performing temporal clustering and classification to name a few of them. In this paper, the major objective is to find temporally outlier patterns from the temporal database of disjoint time-stamped transactions in a single database scan without a need to scan the database multiple times. We aim to eliminate n-1 database scans, performed when we use conventional approach of finding temporal patterns. We demonstrate the approach using a case study. The results show that the proposed approach is computationally efficient which is essentially because of single scan performed.

A novel fuzzy similarity measure and prevalence estimation approach for similarity profiled temporal association pattern mining

Abstract Data generated from Sensors, IoT environment and many real time applications is mainly spatial, temporal, or spatio-temporal. Some of them include data generated from geospatial, geographical, medical, weather, finance and environmental applications. Such data objects changes over time. Conventional knowledge discovery techniques available do not address the need for analyzing such complex datasets and hence data analysis has become increasingly complex and challenging. Soft computing principles such as fuzzy logic, evolutionary and nature inspired computations may be applied to analyze dynamically varying data. Analyzing temporal trends of association patterns requires handling the temporal data, as prevalence values of temporal patterns are implicitly vectors. Finding Prevalence values of temporal association patterns and validating them for similarity using conventional approach increases the computational complexity. This makes it challenging as the conventional data mining algorithms do not address this need. In this research, we propose a novel approach for estimation of temporal association pattern prevalence values and a novel temporal fuzzy similarity measure which holds monotonicity to find similarity between any two temporal patterns. Experiments are performed considering naive, sequential, spamine and proposed approach. The results obtained show the proposed approach is promising and reduces computational complexity in terms of computing true prevalence and optimizing execution times.

SRIHASS - a similarity measure for discovery of hidden time profiled temporal associations

Mining and visualization of time profiled temporal associations is an important research problem that is not addressed in a wider perspective and is understudied. Visual analysis of time profiled temporal associations helps to better understand hidden seasonal, emerging, and diminishing temporal trends. The pioneering work by Yoo and Shashi Sekhar termed as SPAMINE applied the Euclidean distance measure. Following their research, subsequent studies were only restricted to the use of Euclidean distance. However, with an increase in the number of time slots, the dimensionality of a prevalence time sequence of temporal association, also increases, and this high dimensionality makes the Euclidean distance not suitable for the higher dimensions. Some of our previous studies, proposed Gaussian based dissimilarity measures and prevalence estimation approaches to discover time profiled temporal associations. To the best of our knowledge, there is no research that has addressed a similarity measure which is based on the standard score and normal probability to find the similarity between temporal patterns in z-space and retains monotonicity. Our research is pioneering work in this direction. This research has three contributions. First, we introduce a novel similarity (or dissimilarity) measure, SRIHASS to find the similarity between temporal associations. The basic idea behind the design of dissimilarity measure is to transform support values of temporal associations onto z-space and then obtain probability sequences of temporal associations using a normal distribution chart. The dissimilarity measure uses these probability sequences to estimate the similarity between patterns in z-space. The second contribution is the prevalence bound estimation approach. Finally, we give the algorithm for time profiled associating mining called Z-SPAMINE that is primarily inspired from SPAMINE. Experiment results prove that our approach, Z-SPAMINE is computationally more efficient and scalable compared to existing approaches such as Naïve, Sequential and SPAMINE that applies the Euclidean distance.

A similarity measure for temporal pattern discovery in time series data generated by IoT

Internet of Things implicitly generates myriads of temporal data. Unlocking such temporal data becomes a huge concern. Discovery and prediction of repeating temporal patterns and understanding the underlying temporal trends is much more challenging in the case of time stamped temporal data. At present, existing approaches do not reveal seasonal patterns, emerging or diminishing patterns. Determining similar temporal patterns and unearthing eccentric patterns require an efficient dissimilarity measure. This research addresses the similarity measure for revealing similar temporal patterns from time series data generated by IoT.

An Approach for Mining Similar Temporal Association Patterns in Single Database Scan

Mining similar temporal association patterns from a time stamped temporal database is an important research problem in temporal data mining. The main objective and idea of this research is in finding similar temporal patterns from a given time stamped temporal database of transactions by scanning the input database only once. This objective to find temporally similar patterns through single scan of database coins out an important challenge to devise a single database scan procedure which shall use only support values of items computed in the first database scan, so as to discover all other temporal patterns. In the current research, we come out with a novel procedure to discover similar temporal patterns with respect to a reference sequence of support values for a given threshold limit. In this paper, we propose a novel approach to find similar temporal patterns followed by a case study. The approach is efficient in terms of space and time as it eliminates repeated scan of database by computing temporal frequent patterns or temporally similar patterns in only a single database scan.

ASTRA - A Novel interest measure for unearthing latent temporal associations and trends through extending basic gaussian membership function

Time profiled association mining is one of the important and challenging research problems that is relatively less addressed. Time profiled association mining has two main challenges that must be addressed. These include addressing i) dissimilarity measure that also holds monotonicity property and can efficiently prune itemset associations ii) approaches for estimating prevalence values of itemset associations over time. The pioneering research that addressed time profiled association mining is by J.S. Yoo using Euclidean distance. It is widely known fact that this distance measure suffers from high dimensionality. Given a time stamped transaction database, time profiled association mining refers to the discovery of underlying and hidden time profiled itemset associations whose true prevalence variations are similar as the user query sequence under subset constraints that include i) allowable dissimilarity value ii) a reference query time sequence iii) dissimilarity function that can find degree of similarity between a temporal itemset and reference. In this paper, we propose a novel dissimilarity measure whose design is a function of product based gaussian membership function through extending the similarity function proposed in our earlier research (G-Spamine). Our approach, MASTER (Mining of Similar Temporal Associations) which is primarily inspired from SPAMINE uses the dissimilarity measure proposed in this paper and support bound estimation approach proposed in our earlier research. Expression for computation of distance bounds of temporal patterns are designed considering the proposed measure and support estimation approach. Experiments are performed by considering naïve, sequential, Spamine and G-Spamine approaches under various test case considerations that study the scalability and computational performance of the proposed approach. Experimental results prove the scalability and efficiency of the proposed approach. The correctness and completeness of proposed approach is also proved analytically.

A Computationally Efficient Approach for Mining Similar Temporal Patterns

Temporal association patterns are those patterns which are obtained from time stamped temporal databases. A temporal association pattern is said to be similar if it satisfies specified subset constraints. The apriori algorithm which is designed for static databases cannot be extended to find similar temporal patterns from temporal databases as patterns are vectors with supports computed at different time slots and Euclidean distance do not satisfy monotonicity property. The brute force approach to find similar temporal patterns requires computing \(2^n\) true support combinations for ‘n’ items from finite item set and problem falls in NP-class. In this present research, we come up with novel approach to discover temporal association which are similar for pre-specified subset constraints, and substantially reduce support computations. The proposed approach eliminates computational overhead in finding similar temporal patterns. The results prove that the proposed method outperforms brute force approach.