Results 11 - 20
of
30
Vijayakumar.V et al. / International Journal on Computer Science and Engineering (IJCSE) Mining Best-N Frequent Patterns in a Video Sequence
"... ABSTRACT — Video mining is used to discover and describe interesting patterns in video data, which has become one of the core problem areas of the data mining research community. Compared to the mining of other types of data (e.g., text), video mining is still in its infancy, and an under-explored f ..."
Abstract
- Add to MetaCart
(Show Context)
ABSTRACT — Video mining is used to discover and describe interesting patterns in video data, which has become one of the core problem areas of the data mining research community. Compared to the mining of other types of data (e.g., text), video mining is still in its infancy, and an under-explored field. There are many challenging research problems facing video mining. Video Association Mining is a relatively new and emerging research trend. It consists two key phases are (i) Video pre-processing and (ii) Frequent Temporal Pattern Mining. The first phase converts the original input video to a sequence format. The second phase concerns the generation of frequent patterns. Frequent pattern generation plays an essential role in mining of association rules. The usual framework is to use a minimal support threshold to obtain all frequent patterns. However, it is nontrivial for users to choose a suitable minimal support threshold. The paper addresses the issue of frequent temporal pattern mining and studies algorithms for the same. In this paper, we proposed a new mining task called mining Best-N frequent patterns, where N is the largest rank value of all frequent patterns to be mined. An efficient algorithm called Modified VidApriori is used to mining Best-N frequent patterns. During the mining process, the undesired patterns are filtered and useful patterns are selected to generate other longer potential frequent patterns. This strategy greatly reduces the search space. The existing Apriori based algorithm is compared with Modified VidApriori. We also presented results of applying these algorithms to a synthetic data set, which show the effectiveness of our algorithm.
Utilizing Non-Redundant Association Rules from Multi-Level Datasets
"... Association rule mining and recommender systems are two popular methods for obtaining knowledge and information from datasets. However, both of these methods suffer from limitations. Traditionally association rule mining has focused on extracting as many rules as possible from flat datasets. More re ..."
Abstract
- Add to MetaCart
(Show Context)
Association rule mining and recommender systems are two popular methods for obtaining knowledge and information from datasets. However, both of these methods suffer from limitations. Traditionally association rule mining has focused on extracting as many rules as possible from flat datasets. More recently, issues over the number of rules and obtaining rules from datasets with multiple concept levels have come into focus. Recommender systems have been popular with users when it comes to helping find similar interests to those they already have. However, recommender systems suffer from two major problems, cold start and novelty. The aims of our research is to develop an approach for extracting non-redundant multi-level and crosslevel association rules from datasets with multiple concept levels and utilise them in a recommender system with the aim of potentially solving the cold start and novelty problems. 1.
An Efficient Technique for Indexing Temporal Databases
"... Temporal databases added a new dimension to traditional transaction databases. This dimension is the life time of each item, i.e. exhibition period, starting from the partition when this item appears in the transaction database to the partition when this item no longer exists. Mining temporal associ ..."
Abstract
- Add to MetaCart
(Show Context)
Temporal databases added a new dimension to traditional transaction databases. This dimension is the life time of each item, i.e. exhibition period, starting from the partition when this item appears in the transaction database to the partition when this item no longer exists. Mining temporal association rules became very interesting topic in many applications nowadays. In this paper, an efficient technique is proposed for indexing temporal databases in order to facilitate support counting process during mining operation. Some experiments were conducted using well-known real datasets to show the performance of the proposed indexing technique with respect to index size and running time of the mining algorithm. The results show that the proposed indexing technique saves a lot of running time and works efficiently with different databases characteristics.
Sunder Deep Group of
"... Association rule mining is a way to find relations or co-relations among a set of information available. The aim to generate rules for giving multiple data from various databases. Analysis of data can be possible with the help of sequential access of data from database. In case of sequential access ..."
Abstract
- Add to MetaCart
(Show Context)
Association rule mining is a way to find relations or co-relations among a set of information available. The aim to generate rules for giving multiple data from various databases. Analysis of data can be possible with the help of sequential access of data from database. In case of sequential access of data it may cause multiple times same rules to be generated. It is desired to find a solution to get out of those unnecessary association rules due to the complex characteristics of serial data. Although many numbers of serial association rule with the use of either sequence or temporal constraint as prediction model, these two models did not consider with the repetition during the process of rule mining for the database. The goal of this paper is to propose a method for redundancy free serial association rule mining.
EFFICIENT APPROACH TO DISCOVER INTERVAL-BASED SEQUENTIAL PATTERNS
"... In most of the sequential pattern mining methodology they have concentrated only on time point base event data. But some research efforts have detailed the mining patterns from time interval based event data. In many application most of the events are occurred at time interval based event not a poin ..."
Abstract
- Add to MetaCart
(Show Context)
In most of the sequential pattern mining methodology they have concentrated only on time point base event data. But some research efforts have detailed the mining patterns from time interval based event data. In many application most of the events are occurred at time interval based event not a point based interval for example patient affected by the certain time period. Our goal is to mine the frequently occurred sequential patterns in the database. In this study we have introduced a new algorithm namely KPrefixspan by modifying the TPrefixspan algorithm to overcome the demerits of that algorithm. Here new approach called refined database can reduce the scanning time extremely since the unsupported events are removed at each projection also result of the sequential pattern is extremely precise. Experiments constructed for synthetic datasets. From the experimental results we reduced the running time almost 60 % and also reduce the memory usage almost 25 % when compared to the existing TPrefixspan algorithm.
An Overview on Mobile Data Mining
"... In early days the mobile phones are considered to perform only telecommunication operation. This scenario of mobile phones as communication devices changed with the emergence of a new class of mobile devices called the smart phones. These smart phones in addition to being used as a communication dev ..."
Abstract
- Add to MetaCart
In early days the mobile phones are considered to perform only telecommunication operation. This scenario of mobile phones as communication devices changed with the emergence of a new class of mobile devices called the smart phones. These smart phones in addition to being used as a communication device are capable of doing things that a computer does. In recent times the smart phone are becoming more and more powerful in both computing and storage aspects. The data generated by the smart phone provide a means to get new knowledge about various aspects like usage, movement of the user etc. This paper provides an introduction to Mobile Data Mining and its types.
Mining Data Using Various Association Rule Mining Algorithms in Distributed Environment Using MPI
"... Abstract- Data mining combines machine learning, statistics and visualization techniques to discover and extract knowledge. In order to improve the efficiency of mining algorithm for the large data sets we are implementing Distributed Data Mining (DDM). In distributed association rule mining algorit ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract- Data mining combines machine learning, statistics and visualization techniques to discover and extract knowledge. In order to improve the efficiency of mining algorithm for the large data sets we are implementing Distributed Data Mining (DDM). In distributed association rule mining algorithm, one of the major challenges is to reduce the communication overhead. Data sites are required to exchange lot of information in the data mining process which may generates massive communication overhead. Message passing interface (MPI) is a technique to exchange information among a number of communicating nodes. Here we apply association rule mining algorithms like TopKRules and TNR algorithm in distributed environment using MPI for mining data within less communication overhead.
Graph Based Approach for Finding Frequent Itemsets to Discover Association Rules
"... Abstract — The discovery of association rules is an important task in data mining and knowledge discovery. Several algorithms have been developed for finding frequent itemsets and mining comprehensive association rules from the databases. The efficiency of these algorithms is a major issue since a l ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract — The discovery of association rules is an important task in data mining and knowledge discovery. Several algorithms have been developed for finding frequent itemsets and mining comprehensive association rules from the databases. The efficiency of these algorithms is a major issue since a long time and has captured the interest of a large community of researchers. This paper presents a new approach that can mine frequent itemset or patterns in less time and in a straight forward way. Majority of the algorithms developed for finding frequent itemsets scan the database repeatedly and are based on the concept of minimum threshold support value. The proposed approach is based on graph and finds frequent itemsets without repeatedly scanning the database. The algorithm finds frequent itemsets irrespective of support level and can be used for finding the largest most frequent itemset. The proposed approach performs single scan of the database in first phase and draws a graph in which edges are labeled with the respective transactions ids. In second phase, a table is constructed which has all distinct labels and the corresponding itemsets. The largest most frequent itemset is selected from this table according to the given selection criterion. Index Terms — data mining, association rules, knowledge and data management, database I.
