A Spark ML driven preprocessing approach for deep learning based scholarly data applications

by   Samiya Khan, et al.

Big data has found applications in multiple domains. One of the largest sources of textual big data is scientific documents and papers. Big scholarly data have been used in numerous ways to create innovative applications such as collaborator discovery, expert finding and research management systems. With the advent of advanced machine and deep learning techniques, the accuracy and novelty of such applications have risen manifold. However, the biggest challenge in the development of deep learning models for scholarly applications in cloud based environment is the underutilization of resources because of the excessive time taken by textual preprocessing. This paper presents a preprocessing pipeline that makes use of Spark for data ingestion and Spark ML for pipelining preprocessing tasks. The evaluation of the proposed work is done using a case study, which uses LSTM based text summarization for generating title or summary from abstract of any research. The ingestion, preprocessing and cumulative time for the proposed approach (P3SAPP) is much lower than the conventional approach (CA), which manifests in reduction of costs as well.



There are no comments yet.



DPASF: A Flink Library for Streaming Data preprocessing

Data preprocessing techniques are devoted to correct or alleviate errors...

Big Machinery Data Preprocessing Methodology for Data-Driven Models in Prognostics and Health Management

Sensor monitoring networks and advances in big data analytics have guide...

nuts-flow/ml: data pre-processing for deep learning

Data preprocessing is a fundamental part of any machine learning applica...

Toward Automated Website Classification by Deep Learning

In recent years, the interest in Big Data sources has been steadily grow...

ContainerStress: Autonomous Cloud-Node Scoping Framework for Big-Data ML Use Cases

Deploying big-data Machine Learning (ML) services in a cloud environment...

Robust Data Preprocessing for Machine-Learning-Based Disk Failure Prediction in Cloud Production Environments

To provide proactive fault tolerance for modern cloud data centers, exte...

Deep recommender engine based on efficient product embeddings neural pipeline

Predictive analytics systems are currently one of the most important are...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.