Md. Samiullah

Effective periodic pattern mining in time series databases

The goal of analyzing a time series database is to find whether and how frequent a periodic pattern is repeated within the series. Periodic pattern mining is the problem that regards temporal regularity. However, most of the existing algorithms have a major limitation in mining interesting patterns of users interest, that is, they can mine patterns of specific length with all the events sequentially one after another in exact positions within this pattern. Though there are certain scenarios where a pattern can be flexible, that is, it may be interesting and can be mined by neglecting any number of unimportant events in between important events with variable length of the pattern. Moreover, existing algorithms can detect only specific type of periodicity in various time series databases and require the interaction from user to determine periodicity. In this paper, we have proposed an algorithm for the periodic pattern mining in time series databases which does not rely on the user for the period value or period type of the pattern and can detect all types of periodic patterns at the same time, indeed these flexibilities are missing in existing algorithms. The proposed algorithm facilitates the user to generate different kinds of patterns by skipping intermediate events in a time series database and find out the periodicity of the patterns within the database. It is an improvement over the generating pattern using suffix tree, because suffix tree based algorithms have weakness in this particular area of pattern generation. Comparing with the existing algorithms, the proposed algorithm improves generating different kinds of interesting patterns and detects whether the generated pattern is periodic or not. We have tested the performance of our algorithm on both synthetic and real life data from different domains and found a large number of interesting event sequences which were missing in existing algorithms and the proposed algorithm was efficient enough in generating and detecting periodicity of flexible patterns on both types of data.

Queue management based congestion control in wireless body sensor network

We consider the issue of designing a transport layer protocol for energy efficient congestion control and reliable data transfer in wireless body sensor network. The proposed protocol focuses on efficient management of queue to provide reliability and reduce packet loss. This protocol achieves energy efficiency to a greater extent by reducing packet loss. The protocol achieved greater throughput by ensuring reliability in the network. It is capable of supporting multiple applications in the same network by introducing unique packet sequence number. We present the design and implementation of the protocol and evaluate the protocol with different scenarios and network characteristics.

An efficient approach to mine flexible periodic patterns in time series databases

Periodic pattern mining in time series databases is one of the most interesting data mining problems that is frequently appeared in many real-life applications. Some of the existing approaches find fixed length periodic patterns by using suffix tree structure, i.e., unable to mine flexible patterns. One of the existing approaches generates periodic patterns by skipping intermediate events, i.e., flexible patterns, using apriori based sequential pattern mining approach. Since, apriori based approaches suffer from the issues of huge amount of candidate generation and large percentage of false pattern pruning, we propose an efficient algorithm FPPM (Flexible Periodic Pattern Mining) using suffix trie data structure. The proposed algorithm can capture more effective variable length flexible periodic patterns by neglecting unimportant or undesired events and considering only the important events in an efficient way. To the best of our knowledge, ours is the first approach that simultaneously handles various starting position throughout the sequences, flexibility among events in the mined patterns and interactive tuning of period values on the go. Complexity analysis of the proposed approach and comparison with existing approaches along with analytical comparison on various issues have been performed. As well as extensive experimental analyses are conducted to evaluate the performance of proposed FPPM algorithm using real-life datasets. The proposed approach outperforms existing algorithms in terms of processing time, scalability, and quality of mined patterns. HighlightsDevised a new algorithm to generate flexible periodic patterns using suffix trie.Handling variable starting position for mining periodicity without recalculation.A new periodicity detection system to find more interesting periodic patterns.Mining periodicity in a single run and database scan in more interactive manner.Efficiency and scalability of proposed approach are tested with real life datasets.

An efficient approach for mining cross-level closed itemsets and minimal association rules using closed itemset lattices

Multilevel knowledge in transactional databases plays a significant role in our real-life market basket analysis. Many researchers have mined the hierarchical association rules and thus proposed various approaches. However, some of the existing approaches produce many multilevel and cross-level association rules that fail to convey quality information. From these large number of redundant association rules, it is extremely difficult to extract any meaningful information. There also exist some approaches that mine minimal association rules, but these have many shortcomings due to their naive-based approaches. In this paper, we have focused on the need for generating hierarchical minimal rules that provide maximal information. An algorithm has been proposed to derive minimal multilevel association rules and cross-level association rules. Our work has made significant contributions in mining the minimal cross-level association rules, which express the mixed relationship between the generalized and specialized view of the transaction itemsets. We are the first to design an efficient algorithm using a closed itemset lattice-based approach, which can mine the most relevant minimal cross-level association rules. The parent-child relationship of the lattices has been exploited while mining cross-level closed itemset lattices. We have extensively evaluated our proposed algorithms efficiency using a variety of real-life datasets and performing a large number of experiments. The proposed algorithm has outperformed the existing related work significantly during the pervasive performance comparison.

Mining interesting patterns from uncertain databases

We proposed a strategy for weighted uncertain interesting pattern mining.It computes expected support confidence to mine weighted correlated patterns.It prunes infrequent patterns early by computing prefix proxy values.It constructs a tree to capture prefix cap & proxy values for uncertain databases.Our strategy generates a manageable number of interesting patterns quickly. Due to a growing demand for efficient algorithms for mining frequent itemsets from uncertain databases, several approaches have been proposed in recent years, but all of them use support-based constraints to prune the combinatorial search space. Most real life databases contain data whose correctness is uncertain. The support-based constraint alone is not enough, because the frequent itemsets may have weak affinity. Even a very high minimum support is not effective for finding correlated patterns with increased weight or support affinity. There are a few approaches in precise databases that propose new measures to mine correlated patterns, but they are not applicable in uncertain databases because certain and uncertain databases differ both semantically and computationally. In this paper, we propose a new strategy: Weighted Uncertain Interesting Pattern Mining (WUIPM), in which a tree structure (WUIP-tree) and several new measures (e.g., uConf, wUConf) are suggested to mine correlated patterns from uncertain databases. To our knowledge, ours is the first work specifically to consider weight or importance of an individual item alongside correlation between items of patterns in uncertain databases. Additionally, we propose a new metric, prefix proxy value, pProxy for our WUIP-tree that helps improve the mining performance. A comprehensive performance study shows that our strategy (a) generates fewer but valuable patterns and (b) is faster than existing approaches even when affinity measures are not applied.

A new framework for mining frequent interaction patterns from meeting databases

Meetings play an important role in workplace dynamics in modern life since their atomic components represent the interactions among human beings. Semantic knowledge can be acquired by discovering interaction patterns from these meetings. A recent method represents meeting interactions using tree data structure and mines interaction patterns from it. However, such a tree based method may not be able to capture all kinds of triggering relations among interactions and distinguish same interaction from different participants of different ranks. Hence, it is not suitable to find all interaction patterns such as those about correlated interactions. In this paper, we propose a new framework for mining interaction patterns from meetings using an alternative data structure, namely, weighted interaction flow directed acyclic graph (WIFDAG). Specifically, a WIFDAG captures both temporal and triggering relations among interactions in meetings. Additionally, to distinguish participants from different ranks, we assign weights to nodes in the WIFDAGs. Moreover, we also propose an algorithm called WDAGMeet for mining weighted frequent interaction patterns from meetings represented by the proposed framework. Extensive experimental results are shown to signify the effectiveness of the proposed framework and the mining algorithm built on that framework for mining frequent interaction patterns from meetings. HighlightsWe proposed a DAG-based mining framework to model and mine interactions in meetings.The framework integrates DAG, interaction pattern & weighted frequent pattern miningIt captures temporal and triggering relations among meeting interactions.It incorporates node weight to preserve rank information of meeting participants.It exploits anti-monotone property and is practical in many real-life scenarios.

A new framework for mining weighted periodic patterns in time series databases

Developing a new weight-based framework for periodic pattern mining.Devising an efficient weighted periodic pattern mining algorithm with suffix trie.Different pruning strategies are introduced to accelerate the performance.Capable of mining symbol, partial, full-cycle periodicity in a single run.The results on real datasets show efficiency and effectiveness of our approach. Mining periodic patterns in time series databases is a daunting research task that plays a significant role at decision making in real life applications. There are many algorithms for mining periodic patterns in time series, where all patterns are considered as uniformly same. However, in real life applications, such as market basket analysis, gene analysis and network fault experiment, different types of items are found with several levels of importance. Again, the existing algorithms generate huge periodic patterns in dense databases or in low minimum support, where most of the patterns are not important enough to participate in decision making. Hence, a pruning mechanism is essential to reduce these unimportant patterns. As a purpose of mining only important patterns in a minimal time period, we propose a weight based framework by assigning different weights to different items. Moreover, we develop a novel algorithm, WPPM (Weighted Periodic Pattern Mining Algorithm), in time series databases underlying suffix trie structure. To the best of our knowledge, ours is the first proposal that can mine three types of weighted periodic pattern, (i.e. single, partial, full) in a single run. A pruning method is introduced by following downward property, with respect of the maximum weight of a given database, to discard unimportant patterns. The proposed algorithm presents flexibility to user by providing intermediate unimportant pattern skipping opportunity and setting different starting positions in the time series sequence. The performance of our proposed algorithm is evaluated on real life datasets by varying different parameters. At the same time, a comparison between the proposed and an existing algorithm is shown, where the proposed approach outperformed the existing algorithm in terms of time and pattern generation.

Mining frequent correlated graphs with a new measure

Correlation mining is recognized as one of the most important data mining tasks for its capability to identify underlying dependencies between objects. On the other hand, graph-based data mining techniques are increasingly applied to handle large datasets due to their capability of modeling various non-traditional domains representing real-life complex scenarios such as social/computer networks, map/spatial databases, chemical-informatics domain, bio-informatics, image processing and machine learning. To extract useful knowledge from large amount of spurious patterns, correlation measures are used. Nonetheless, existing graph based correlation mining approaches are unable to capture effective correlations in graph databases. Hence, we have concentrated on graph correlation mining and proposed a new graph correlation measure, gConfidence, to discover more useful graph patterns. Moreover, we have developed an efficient algorithm, CGM (Correlated Graph Mining), to find the correlated graphs in graph databases. The performance of our scheme was extensively analyzed in several real-life and synthetic databases based on runtime and memory consumption, then compared with existing graph correlation mining algorithms, which proved that CGM is scalable with respect to required processing time and memory consumption and outperforms existing approaches by a factor of two in speed of mining correlations.

Movie swarm: Information mining technique for movie recommendation system

Huge amount of movies are available over the world, all of those are impossible to see for one user and all of them are not interesting. Movie recommendation systems filter out irrelevant movies and suggest the relevant movies those would be interesting for users. Traditional system can not recommend new users and new items efficiently. In collaborative filtering recommendation is based on users activities and products features hence when new users enter the system and new items added, it can not recommend. In content based recommend can recommend new items based on items features but unable to recommend new users. Therefor, We have proposed an information mining tool that collect all important information which is needed in movie recommendation system. In our proposed system, we have generated movie swarm which is very useful for movie producers and can solve new items problem. Also finds out which genres of movie should be recommended among followers, that solves new users recommendation problem. Experimental studies on the real data demonstrate the encourages and effectiveness of our methods.

Supergraph based periodic pattern mining in dynamic social networks

In dynamic networks, periodically occurring interactions express especially significant meaning. However, these patterns also could occur infrequently, which is why it is difficult to detect while working with mass data. To identify such periodic patterns in dynamic networks, we propose single pass supergraph based periodic pattern mining SPPMiner technique that is polynomial unlike most graph mining problems. The proposed technique stores all entities in dynamic networks only once and calculate common sub-patterns once at each timestamps. In this way, it works faster. The performance study shows that SPPMiner method is time and memory efficient compared to others. In fact, the memory efficiency of our approach does not depend on dynamic network’s lifetime. By studying the growth of periodic patterns in social networks, the proposed research has potential implications for behavior prediction of intellectual communities.