Data stream mining to address big data problems

Olmezogullari, Erdi; Arı, İsmail; Celebi, Omer Faruk; Ergüt, Salih

doi:10.1109/siu.2013.6531483

Cited by 2 publications

(2 citation statements)

References 8 publications

(2 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To mention a few of them, there are works aimed at learning classifier chains for data mining [37], improving routing and navigation services [38], recommending musical preferences [39], detecting trends on Twitter and Bitly [40], filtering spam [41], processing image and video streams [42], analyzing streams of aviation data [43], monitoring underwater acoustics [44], discovering communities in Twitter during natural disasters [45], detecting issues in electric power systems [46], searching in logs [47] or summarizing microblogging data [48], among others. The latter four of the forementioned works use HBase and Hadoop, proving it to be a convenient storage system for performing streaming analysis; and some of the previous works also provide their products as a service, so that the real time analytics system can be accessed through a REST API.…”

Section: State Of the Artmentioning

confidence: 99%

A scalable machine learning online service for big data real-time analysis

Baldominos

Albacete

Sáez

et al. 2014

2014 IEEE Symposium on Computational Intelligence in Big Data (CIBD)

View full text Add to dashboard Cite

Abstract-This work describes a proposal for developing and testing a scalable machine learning architecture able to provide real-time predictions or analytics as a service over domain-independent big data, working on top of the Hadoop ecosystem and providing real-time analytics as a service through a RESTful API. Systems implementing this architecture could provide companies with on-demand tools facilitating the tasks of storing, analyzing, understanding and reacting to their data, either in batch or stream fashion; and could turn into a valuable asset for improving the business performance and be a key market differentiator in this fast pace environment. In order to validate the proposed architecture, two systems are developed, each one providing classical machine-learning services in different domains: the first one involves a recommender system for web advertising, while the second consists in a prediction system which learns from gamers' behavior and tries to predict future events such as purchases or churning. An evaluation is carried out on these systems, and results show how both services are able to provide fast responses even when a number of concurrent requests are made, and in the particular case of the second system, results clearly prove that computed predictions significantly outperform those obtained if random guess was used.

show abstract

Section: State Of the Artmentioning

confidence: 99%

A scalable machine learning online service for big data real-time analysis

Baldominos

Albacete

Sáez

et al. 2014

2014 IEEE Symposium on Computational Intelligence in Big Data (CIBD)

View full text Add to dashboard Cite

show abstract

“…Itintroduces the weight characteristics into the FP-growth mining algorithm, making it more suitable for the calculation process of the time FP-stream. Ölmezogulları et al [17] added Apriori and FP-growth algorithms for stream mining inside an event processing engine. The FP-stream [18] is one of the algorithms that dynamically updates frequent item sets with the incoming data streams.…”

Section: Introductionmentioning

confidence: 99%

A Caching Strategy for Transparent Computing Server Side Based on Data Block Relevance

et al. 2018

View full text Add to dashboard Cite

The performance bottleneck of transparent computing (TC) is on the server side. Caching is one of the key factors of the server-side performance. The count of disk input/output (I/O) can be reduced if multiple data blocks that are correlated with the data block currently accessed are prefetched in advance. As a result, the service performance and user experience quality in TC can be improved. In this study, we propose a caching strategy for the TC server side based on data block relevance, which is called the correlation pattern-based caching strategy (CPCS). In this method, we adjust a model that is based on a frequent pattern tree (FP-tree) for mining frequent patterns from data streams (FP-stream) to the characteristics of data access in TC, and we devise a cache structure in accordance with the storage model of TC. Finally, the access process in TC with real access traces in different caching strategies. Simulation results show that the cache hit rate under the CPCS is higher than that using other algorithms under conditions in which the parameters are coordinated properly.

show abstract

Data stream mining to address big data problems

Cited by 2 publications

References 8 publications

A scalable machine learning online service for big data real-time analysis

A scalable machine learning online service for big data real-time analysis

A Caching Strategy for Transparent Computing Server Side Based on Data Block Relevance

Contact Info

Product

Resources

About