Deep Learning Approaches for Big Data Analytics: Opportunities, Issues and Research Directions (26-33) (original) (raw)

A survey on deep learning in big data analytics

Industry 4.0, 2020

Over the last few years, Deep learning has begun to play an important role in analytics solutions of big data. Deep learning is one of the most active research fields in machine learning community. It has gained unprecedented achievements in fields such as computer vision, natural language processing and speech recognition. The ability of deep learning to extract high-level complex abstractions and data examples, especially unsupervised data from large volume data, makes it attractive a valuable tool for big data analytics. In this paper, we review the deep learning architectures which can be used for big data processing. Next, we focus on the analysis and discussions about the challenges and possible solutions of deep learning for big data analytics. Finally, have been outlined several open issues and research trends.

Deep learning applications and challenges in big data analytics

Big Data Analytics and Deep Learning are two high-focus of data science. Big Data has become important as many organizations both public and private have been collecting massive amounts of domain-specific information, which can contain useful information about problems such as national intelligence, cyber security, fraud detection, marketing, and medical informatics. Companies such as Google and Microsoft are analyzing large volumes of data for business analysis and decisions, impacting existing and future technology. Deep Learning algorithms extract high-level, complex abstractions as data representations through a hierarchical learning process. Complex abstractions are learnt at a given level based on relatively simpler abstractions formulated in the preceding level in the hierarchy. A key benefit of Deep Learning is the analysis and learning of massive amounts of unsupervised data, making it a valuable tool for Big Data Analytics where raw data is largely unlabeled and un-categorized. In the present study, we explore how Deep Learning can be utilized for addressing some important problems in Big Data Analytics, including extracting complex patterns from massive volumes of data, semantic indexing, data tagging, fast information retrieval, and simplifying discriminative tasks. We also investigate some aspects of Deep Learning research that need further exploration to incorporate specific challenges introduced by Big Data Analytics, including streaming data, high-dimensional data, scalability of models, and distributed computing. We conclude by presenting insights into relevant future works by posing some questions, including defining data sampling criteria, domain adaptation modeling, defining criteria for obtaining useful data abstractions, improving semantic indexing, semi-supervised learning, and active learning.

Big Data Deep Learning: Challenges and Perspectives

Deep learning is currently an extremely active research area in machine learning and pattern recognition society. It has gained huge successes in a broad area of applications such as speech recognition, computer vision, and natural language processing. With the sheer size of data available today, big data brings big opportunities and transformative potential for various sectors; on the other hand, it also presents unprecedented challenges to harnessing data and information. As the data keeps getting bigger, deep learning is coming to play a key role in providing big data predictive analytics solutions. In this paper, we provide a brief overview of deep learning, and highlight current research efforts and the challenges to big data, as well as the future trends.

A REVIEW OF IMPLEMENTATION OF DEEP LEARNING IN BIG DATA ANALYSIS

Data is being generated very rapidly and at very high magnitudes. This data, termed as 'Big Data' has found its use in many fields. The aim of this paper is to discuss the challenges posed by Big Data analysis and the techniques that can be used to solve these challenges. One of the most efficient techniques used to do so is Deep Learning. This paper also focuses on the models that have been implemented for big data analysis with deep learning and its applications.

OPPORTUNITIES AND GAINSAYS IN DEEP LEARNING ON BIG DATA

Deep learning is currently an extremely active research area in pattern recognition society. It has gained huge successes in a broad area of applications such as speech recognition, computer vision, and natural language processing. With the sheer size of data available today, big data brings big opportunities and transformative potential for various sectors; on the other hand, it also presents unprecedented challenges to harnessing data and information. As the data keeps getting bigger, Deep learning is coming to play a key role in providing big data predictive analytics solutions.Big data assist ML algorithms to uncover more fine-grained patterns and helps in accurate predictions .The major challenges to ML are model scalability and distributed computing.The realization of this grand potential relies on the ability to extract value from such massive data through data analytics; Deep learning is at its core because of its ability to learn from data and provide data driven insights, decisions, and predictions. In this paper first, we review the Deep learning techniques and highlight some promising learning methods in recent studies, such as representation learning, deep learning, distributed and parallel learning, transfer learning, active learning, and kernel-based learning and analyse the challenges and possible solutions of Deep learning for big data. Finally, we outline several open issues and research trends.

DEEP LEARNING BASED ON BIG DATA ANALYTICS

IAEME PUBLICATION, 2020

Deep learning techniques are widely extended to numerous scientific disciplines and information technology like voice recognition, object definitions, and learning processes in visual processing. Likewise, conventional data analysis methods have many constraints of processing massive amount of information. Deep Learning is actually a rather emerging field in neural networks and culture of artificial intelligence. It has gained enormous success in important field technologies like Machine learning, Voice and Video Analysis, and Machine Translation Processing. There are huge quantities of information produced by numerous sources daily. Therefore, the data concept is translated to Analytics that poses difficulties in the phases of knowledge processing and judgment-making. Furthermore, Big Data analytics needs new and advanced methodologies depending on system and deep learning methods to analyze data in real-time with high reliability and productivity. Deep learning skills can promote the handling of such information, particularly their capacity to handle both the marked and unmarked data which are sometimes amply gathered in Big Data. The paper provides a comprehensive of Big Data and discusses particular problems in analytics that Deep Learning can solve.

Deep learning and Big Data Analysis: Challenges, Opportunities and Applications

With the approach of IoT, there are tremendous changes going in the measure of information. and deep learning are the two most critical concentration focuses in this universe of computerized science. Information is considered an indispensible need these days. This structure of open and private information is likewise exceptionally import primary favourable position of DL is looking at and gets out taking in measures from it for unsupervised and managed data, by making it a noteworthy gadget for Big Data investigation where crude information is accumulated and is marked and unlabel long last arranged. In the present paper, we examine with reference to how Deep Learning is exploited for some basic issues in Big Data Analytics, which additionally incorporates extraction of complex information from colossal masses of data requesting, data marking, speedy information recuperation, and enhancing discriminative errands. Huge information has now ended up being extremely famous as a couple of affiliations are gathering huge measures of room specific information which utilized to address the issues related to national learning, computerized security, distortion area, and wellbeing informatics. Profound Learning Big Data permits extraction of not typical condition of information, complex thoughts as data depiction through a different levelled learning process. A key favourable position of Deep Learning is Big Data investigation examination that it can pick up from huge mass of unsupervised data. In this manner, making it an imperative instrument for Big Data Analytics where monstrous measures of data and information are not group. This part weights on the need for immense information, mechanical kinds of @ IJTSRD |

Survey of Deep-Learning Techniques in Big-Data Analytics

Wireless Personal Communications

The present elaboration of Big-data research studies relying upon Deep-learning methods had revitalized the decision-making mechanism in the business sectors and the enterprise domains. The rms' operational parameters also have the dependency of the Big-data analytics phase, their way of managing the data, and to evolve the outcomes of Big-data implementation by using the Deep-learning algorithms. The present enhancements in the Deep-learning approaches in Big-data applications facilitate the decision-making process such as the information-processing to the employees, analytical potentials augmentation, and in the transition to having more innovative work. In this DL-approach, the robustpatterns of the data-predictions resulted from the unstructured information by conceptualizing the Decision-making methods. Hence this paper elaborates the above statements stating the impact of the Deep-learning process utilizing the Big-data to operate in the enterprise and Business sectors. Also this study provides a comprehensive survey of all the Deep-learning techniques illustrating the e ciency of Big-Data processing on having the impacts of operational parameters, concentrating the datadimensionality factors and the Big-data complications rectifying by utilizing the DL-algorithms, usage of Machine-learning or deep-learning process for the decision-making mechanism in the Enterprise sectors and business sectors, the predictions of the Big-data analytics resulting to the decision parameters within the organisations, and in the management of larger scale of datasets in Big-data analytics processing by utilizing the Deep-learning implementations. The comparative analysis of the reviewed studies has also been described by comparing existing approaches of Deep-learning methodologies in employing Big-data analytics. in constructing the applications of Deep-learning in the big-data stream. Similarly, a BigDL-distributed-Data-Deep learning-framework for Apache-framework is presented in work [1] which is utilized by several users of rms implementing the Deep-learning application in the Big-Data platforms of production. This model also permits the Applications of Deep-learning process to execute on the Apache-Hadoop clusters or the spark-clusters. This would allow the processing of Big-Data of production and generate the deployment pipeline and management of data-analysis. This study produces the overview of BigDLdistributed-Data-Deep learning-framework for Apache-framework comprising of the distribution-model of execution, training-scalability, real-time use cases and the performance of the computation process. Such type of e cient Big-Data applications employed by using the Deep-learning techniques. Another application of studies is that the studies also illustrated the deep-learning basis framework of the feature extraction process and aided in building the security-primitives. This has revealed that Auto-encoders to be implemented to transmit the conventional-state variables to the smaller type of dimensions. Another study is presenting the application of deep-learning in the construction of security primitives. This study [2] implements the feature-extraction model of the deep-learning basis framework in constructing the security constraints. It is also shown that the deep-auto-encoders would be utilized in the transformation of state-variables space to the smaller dimensions count, for instance, the power-ows. The study's inferences reveal that the proposed-model is the data-driven approach, and it is utilized In many applications within the security-evaluation context. The framework exhibited a higher level of performance by learning case-studies and comparison methods. Hence the Deep-learning also plays a role in solving the data-dimensionality and big-data complications. The multi-criteria basis decision-making process is also the key factors to overcome the complications associated with Big-data analytics. This processes would opt to determine the solution on the basis of recent machine-learning approaches such as the decision-making processed yielding the Big-data insights. Another dimension of the article focusses on the transition phase of the analytics phase to AIarti cial-Intelligence [3]. The various approaches in evaluating analytical-capabilities and the business strategy's progress, and the rm's plan in the AI stream are brie y discussed in the article. The study illustrates how the AI-stream impacts the enterprise, present capabilities of the business and how the pro cient strategy has to be employed. As the industries and enterprises evolve, the more enormous amount of data, the more vast computing power and e cient speed of the network; hence the manufacturing rms face unorganized dataprocessing issues. But employing the Deep-learning techniques in Big-data analytics, IoT-Internet of Things studies have improvised in data-manipulation of larger datasets. Some of the several present techniques additions to the Deep-learning patterns and the Deep-learning application in the different domains were illustrated. This process of Deep-learning methodology has enhanced sequentially the computing-devices capacity predictions. [4] This is accomplished by the Big-data presence and the aid of superior-learning-model algorithms. Hence as in rectifying solutions to the problems stated above, the reliable performance analysis and the superior, e cient performance of deep-learning processes have grasped the research studies in every eld.

A Survey on Deep Learning in Big Data and its Applications

2021

Individuals can exchange real-time information thanks to the vast spread and reach of social networks. This active participation with the corporate data, as emails, documents, databases, business processor history, etc and content published on the Web, as age and contact details, reviews, comments, photos, images, videos, sounds, texts, famous cookies, or ecommerce transactions, exchanges on social networks, are very important. Data recovery from different sources can be a difficult task. A timely and correct assessment of an event currently under discussion is critical to the effectiveness of the used method. This information, collected in the Web can then be updated. Various ways are developed to automate this necessity, due to the extraction and analysis of correct social media content. Alleviation methods do not adequately incorporate these approaches. It may be necessary to reveal them in order to make further progress, particularly in the areas of energy efficiency and cleaner...