Adam G. M. Pazdor

Parallel social network mining for interesting 'following' patterns

Social networking sites (e.g., Facebook, Google+, and Twitter) have become popular for sharing valuable knowledge and information among social entities (e.g., individual users and organizations), who are often linked by some interdependency such as friendship. As social networking sites keep growing, there are situations in which a user wants to find those frequently followed groups of social entities so that he can follow the same groups. In this article, we present (i) a space‐efficient bitwise data structure for capturing interdependency among social entities; (ii) a time‐efficient data mining algorithm that makes the best use of our proposed data structure for serial discovery of groups of frequently followed social entities; and (iii) another time‐efficient data mining algorithm for concurrent computation and discovery of groups of frequently followed social entities in parallel so as to handle high volumes of social network data. Evaluation results show the efficiency and practicality of our data structure and social network data mining algorithms. Copyright

Knowledge Discovery from Social Graph Data

High volumes of a wide variety of valuable data can be easily collected and generated from a broad range of data sources of different veracities at a high velocity. In the current era of big data, many traditional data management and analytic approaches may not be suitable for handling the big data due to their well-known 5Vs characteristics. Over the past few years, several systems and applications have developed to use cluster, cloud or grid computing to manage and analyze big data so as to support data science (e.g., knowledge discovery and data mining). In this paper, we present a knowledge-based system for social network analysis so as to support big data mining of interesting patterns from big social networks that are represented as graphs.

MapReduce-Based Complex Big Data Analytics over Uncertain and Imprecise Social Networks

With advances in technology, high volumes of valuable but complex data can be easily collected and generated from various sources in the current era of big data. A prime source of these complex big data is the social network, in which users are often linked by some interdependencies such as friendships and follower-followee relationships. These interdependencies can be uncertain and imprecise. Moreover, as the social network keeps growing, there are situations in which individual users or businesses want to find those popular (i.e., frequently followed) groups of users so that they can follow the same groups. In this paper, we present a complex big data analytic solution that uses the MapReduce model to mine uncertain and imprecise social networks for discovering groups of potentially popular users. Evaluation results show the efficiency and practicality of our solution in conducting complex big data analytics over uncertain and imprecise social networks.

Bitwise parallel association rule mining for web page recommendation

For many real-life web applications, web surfers would like to get recommendation on which collections of web pages that would be interested to them or that they should follow. In order to discover this information and make recommendation, data mining---and specially, association rule mining or web mining---is in demand. Since its introduction, association rule mining has drawn attention of many researchers. Consequently, many association rule mining algorithms have been proposed for finding interesting relationships---in the form of association rules---among frequently occurring patterns. These algorithms include level-wise Apriori-based algorithms, tree-based algorithms, hyperlinked array structure based algorithms, and vertical mining algorithms. While these algorithms are popular, they suffer from some drawbacks. Moreover, as we are living in the era of big data, high volumes of a wide variety of valuable data of different veracity collected at a high velocity post another challenges to data science and big data analytics. To deal with these big data while avoiding the drawbacks of existing algorithms, we present a bitwise parallel association rule mining system for web mining and recommendation in this paper. Evaluation results show the effectiveness and practicality of our parallel algorithm---which discovers popular pages on the web, which in turn gives the web surfers recommendation of web pages that might be interested to them---in real-life web applications.

Web Page Recommendation Based on Bitwise Frequent Pattern Mining

In many applications, web surfers would like to get recommendation on which collections of web pages that would be interested to them or that they should follow. In order to discover this information and make recommendation, data mining in general—or frequent pattern mining in specific—can be applicable. Since its introduction, frequent pattern mining has drawn attention from many researchers. Consequently, many frequent pattern mining algorithms have been proposed, which include levelwise Apriori-based algorithms, tree-based algorithms, hyperlinked array structure based algorithms, as well as vertical mining algorithms. While these algorithms are popular, they also suffer from some drawbacks. To avoid these drawbacks, we propose an alternative frequent pattern mining algorithm called BW-mine in this paper. Evaluation results show that our proposed algorithm is both space-and time-efficient. Furthermore, to show the practicality of BW-mine in real-life applications, we apply BW-mine to discover popular pages on the web, which in turn gives the web surfers recommendation of web pages that might be interested to them.

Knowledge Discovery from Big Social Key-Value Data

High volumes of a wide variety of valuable data can be easily collected and generated from a broad range of data sources of different veracities at a high velocity. In the current era of big data, many traditional data management and analytic approaches may not be suitable for handling the big data due to their well-known 5Vs characteristics. Over the past few years, several systems and applications have developed to use cluster, cloud or grid computing to manage and analyze big data so as to support data science (e.g., knowledge discovery and data mining). In this paper, we present a knowledge-based system for social network analysis so as to support big data mining of interesting patterns from big social networks that are represented in the form of key-value pairs.

Big data mining of social networks for friend recommendation

In the current era of big data, high volumes of valuable data can be easily collected and generated. Social networks are examples of generating sources of these big data. Users in these social networks are often linked by some interdependency such as friendship. As these big social networks keep growing, there are situations in which an individual user wants to find popular groups of friends so that he can recommend the same groups to other users. In this paper, we present a big data analytic solution that uses the MapReduce model in mining these big social networks for discovering groups of frequently connected users for friend recommendation. Evaluation results show the efficiency and practicality of our data analytic solution in mining big social networks, discovering popular users, and recommending friends.

Frequent Sequence Mining with Weight Constraints in Uncertain Databases

Pattern mining has drawn attention of researchers because of its high applicability to mine patterns or sequences from probabilistic databases in various real-life applications. Weight of an item, a pattern, or a sequence help data scientists extract interesting information and knowledge for these applications. However, most related works do not handle sequences with weight constraints in uncertain databases. In this paper, we introduce the concept of weighted uncertain sequence mining. We also propose a new algorithm to mine sequences with weight constraints from uncertain databases. The algorithm is applicable for data science tasks like finding changes in fashion trends and forecasting weather or natural calamities. Our evaluation results show the effectiveness of the algorithm and its superiority over the related works.

An Efficient Approach for Mining Frequent Patterns over Uncertain Data Streams

Knowledge discovery in big data is one of most interesting topics in state-of-the-art research, and frequent patterns mining is a major task. With the rapid growth of modern technology, high volumes of data—which are of different veracities (i.e., may be precise or uncertain)—are flowing at a high velocity all over the world. Properties of data temporally changes with changes in the peoples interests, which make the data dynamic. Due to the uncertainty and dynamic properties of data, finding appropriate and efficient approach to ensure the efficient usage of available resources has become a great challenge. In this paper, we design a new memory-efficient data structure, called Uncertain Stream (US)-tree, which stores recent meta-data. We also develop a probabilistic, sliding window based, efficient algorithm—called Uncertain Stream Frequent Pattern (USFP)-growth—for mining frequent patterns from uncertain data streams. Our comprehensive performance evaluation shows that USFP-growth is correct and efficient when compared with recent related approaches.

A Data Science Model for Big Data Analytics of Frequent Patterns

Frequent pattern mining is an important data mining task. Since its introduction, it has drawn attention from many researchers. Consequently, many frequent pattern mining algorithms have been proposed, which include level-wise Apriori-based algorithms, tree-based algorithms, and hyperlinked array structure based algorithms. While these algorithms are popular and benefit from a few advantages, they also suffer from some disadvantages. In the current era of big data, a wide variety of high volumes of valuable data of different veracities can be easily collected and generated at a high velocity. These big data lead to additional challenges for frequent pattern mining. In this paper, we present a data science model for big data analytics of frequent patterns with MapReduce. We evaluated our model by using social networks, which are good examples of big data. Evaluation results show the efficiency and practicality of our data science model in mining and analyzing big data for the discovery of interesting frequent patterns from various real-life applications including social network analysis.