Sentiment Analysis on Hadoop with Hadoop Streaming

Gupta, Piyush; Kumar, Pardeep; Gopal, Girdhar

doi:10.5120/21582-4651

Cited by 6 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach had been noted for its lower accuracy in Mali, Abhyankar, Bhavarthi, Gaidhar, and Bangare (2016); Singh, Piryani, Uddin, and Marisha (2013) and Denecke (2008). Gupta, Kumar, and Gopal (2015) worked on sentiment analysis using Hadoop streaming with Python. The researchers proposed sentiment analysis architecture and used the HDFS for data storage and Map Reduce program for processing.…”

Section: Literature Reviewmentioning

confidence: 99%

Big Data Frameworks for Sites and Products Recommendation

Ogbuju

IfedibaluChukwu

Okonkwo

et al. 2021

Journal of Information

View full text Add to dashboard Cite

The improvement of the IT infrastructure in an e-commerce platform is essential for both customer satisfaction and increased revenue. While different techniques had been applied towards achieving this, there is still need to engage customer feedbacks in providing an all-inclusive solution to the recommendation systems available in the ecommerce domain. The motivation is on making a more exact recommendation with the traditional collaborative system by mining the feedbacks and uncovering their sentiments using big data analytic systems. This paper describes the design of a big data framework that may be used for shopping sites recommendations and another that may be used for product(s) recommendations to prospecting customers. The use of the cross industry standard process for data mining is applied in proposing the new system. Although the techniques of Hadoop/MongoDB tools are described within the proposed designs, it concentrates mainly on the architecture and algorithm of the system in a holistic approach to enable the platform providers, e-commerce merchants and practitioners find a guided implementation of it using any tool of choice.Contribution/Originality: This study documents the use of the cross-industry standard process for data mining methodology to design big data frameworks that implement the application of sentiment analysis of review texts as preprocessed input into the collaborative filtering algorithm for both sites and product recommendation. INTRODUCTIONThere is a global transformation in the economic sector across nations. Electronic or Digital Commerce, commonly known as e-commerce, is thriving globally. The e-commerce businesses have the ability to digitally go outside the country to draw upon customer bases which are definitely larger in the e-commerce context than in a physical retail context. However, setting up a new and compliant IT infrastructure that would effectively handle the demands of the (current and future) e-commerce is always a herculean task. The IT infrastructure is perhaps the most important component in setting up an e-commerce business. The other important components are the merchants of the products/services, the consumers of those products/services and the service providers who deliver the products/services. The IT infrastructure which connects the consumers with the merchants of the products/services must be built on a set of objectives that would meet the needs of the consumers. Top among the needs of the consumers include reducing customer navigation time by offering a smart product recommendation system, providing additional value by profiling, customizing or personalizing offerings, increasing customer loyalty by incentive offerings and receiving customer feedback on the general products/service presentations, image

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Big Data Frameworks for Sites and Products Recommendation

Ogbuju

IfedibaluChukwu

Okonkwo

et al. 2021

Journal of Information

View full text Add to dashboard Cite

show abstract

“…The polarity of negation words was reversed and blind negation words were labelled as negative thus achieving an accuracy of 75%. [16] collected product reviews from Amazon, and used Hadoop's multimode cluster with MapReduce program. Mapper involved parsing files and information extracting, where as the Reducer consisted of stop word removal, polarity reversal of negation words and a polarity calculator to calculate sentiment score of the reviews based on separate dictionaries of positive and negative, and stop words.…”

Section: Related Workmentioning

confidence: 99%

Sentiment Analysis of Product Reviews in the Absence of Labelled Data Using Supervised Learning Approaches

Waqar

Mushtaq

Junejo³

et al. 2020

MJCS

View full text Add to dashboard Cite

With the growing pace of internet usage, there is a vast variety of diverse individual opinions and thoughts avail-able online. Consumer reviews can act as a feedback and as well as a pool of ideas for which they can be of immense importance to any business. With the growth and popularity of opinion-rich resources such as online review sites and personal blogs, people now can and do, actively use information technology to seek out and un-derstand the opinions of others to decide whether to buy a product or not. Social media websites such as Facebook, Twitter, and e-commerce websites such as eBay, Amazon, etc. are being widely used to communicate viewpoints effectively. Assigning a positive or a negative sentiment to these reviews can help companies understand their users and also help users to make better decisions. Sentiment analysis being a challenging task can be tackled using supervised machine learning techniques or through unsupervised lexicon based approaches if labelled data is unavailable. In this study, we show that in absence of labelled product reviews of a particular website, labelled product reviews from a different website can be effectively used to train the supervised techniques to achieve a comparable performance to the unsupervised lexicon based approaches. This approach also benefits by covering all of the product reviews which the lexicon based approaches fail to do so. We deduce this by comparing five su-pervised approached and three lexicon based approaches on iPhone 5s reviews gathered from Amazon, Facebook, and Reevoo blog. Furthermore, we also found that unigram features combined with bigram features give the best results, and the effect of varying the training data size on the performance of ML classifiers in some cases was significant whereas in other cases it did not have any effect. Our results also suggest that reviews from Amazon are easiest to classify, followed by reviews from Reevo, and Facebook reviews are the hardest to classify.

show abstract

“…Before applying sentiment analysis to text various preprocess method was used to clean the text in order to remove noise in the form of slangs, unwanted words. Hadoop mapreduce [12][13][14] is one of the ways of pre-processing text. The process of map reduce involves mapping the input data, shuffling and sorting the mapped data and final step is to reduce the data and store in HDFS.…”

Section: Related Workmentioning

confidence: 99%

Sentiment on Twitter Data Set using Recurrent Neural Network - Long Short Term Memory

2019

IJITEE

View full text Add to dashboard Cite

Social media is a combination of different platforms where a huge amount of user-generated data is collected. People from various parts of the country express their opinions, reviews, feedback and marketing strategies through social media such as Twitter, Facebook, Instagram, and YouTube. It is vital to explore, gather data, analyze them and consolidate the people views for better decision making. Sentiment analysis is a natural language processing for information extraction that identifies the user’s views. It is used for extracting reviews and opinions about the satisfaction of products, the events, and people for understanding the current trends of product or user’s behavior. The paper reviews and analyses the existing general approaches and algorithms for sentiment analysis. The proposed system selected to perform sentiment analysis on Twitter data set is Long Short Term Memory [LSTM] and evaluated with Naive Bayes Approach.

show abstract

Sentiment Analysis on Hadoop with Hadoop Streaming

Cited by 6 publications

References 9 publications

Big Data Frameworks for Sites and Products Recommendation

Big Data Frameworks for Sites and Products Recommendation

Sentiment Analysis of Product Reviews in the Absence of Labelled Data Using Supervised Learning Approaches

Sentiment on Twitter Data Set using Recurrent Neural Network - Long Short Term Memory

Contact Info

Product

Resources

About