A novel adaptable approach for sentiment analysis on big social data

Alaoui, Imane El; Gahi, Youssef; Messoussi, Rochdi; Chaabi, Youness; Todoskoff, Alexis; Kobi, Abdessamad

doi:10.1186/s40537-018-0120-0

Cited by 119 publications

(50 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The successful prediction of positive and negative sentiments, the precision and recall value is improved. From the analysis overall efficiency of the system is determined using f-score value which is computed using eqn (8). The effective extraction of tweets semantic features, audio and visualization features used to determine the word class whether it belongs to positive or negative one.…”

Section: Figure 3: Crmmnlp-recallmentioning

confidence: 99%

See 1 more Smart Citation

Aspect Based Sentiments from Tweets using Co-Ranking Multi-Modal Natural Language Processing Methodologies

Kanipriya*,

Krishnaveni,

Krishnamurthy

et al. 2020

IJRTE

View full text Add to dashboard Cite

Now-a-days people interest to spend their time in social sites especially twitters to post lot of tweets in every day. The posted tweets are used by many users to get the knowledge about the particular applications, products and other search engine queries. With the help of the posted tweets, their emotions and sentiments are derived which are used to get opinion about particular event. Lot of traditional sentiment detection system that has been developed but they failed to analyze huge volume of tweets and online contents with temporal patterns were also difficult to analyze. To overcome the above issues, the co-ranking multi-modal natural language processing based sentiment analysis system was developed to detect the emotions from the posted tweets. Initially, tweets of different events are collected from social sites which are processed by natural language procedures such as Stemming, Lemmatization, Part-of-speech tagging, word segmentation and parsing are applied to get the words related to posted tweets for deriving the sentiments. From the extracted emotions, co-ranking process is applied to get the opinion effectively related to particular event. Then the efficiency of the system is examined using experimental results and discussions. The introduced system recognize the sentiments from tweets with 98.80% of accuracy.

show abstract

Section: Figure 3: Crmmnlp-recallmentioning

confidence: 99%

“…(Imane El Alaoui et al, 2018) [8] Analyzed public opinion from social data using effective sentiment analysis technique. Initially, the public opinion data is collected from large social sites.…”

Section: Introductionmentioning

confidence: 99%

Aspect Based Sentiments from Tweets using Co-Ranking Multi-Modal Natural Language Processing Methodologies

Kanipriya*,

Krishnaveni,

Krishnamurthy

et al. 2020

IJRTE

View full text Add to dashboard Cite

show abstract

“…Ali, Sana, Ahmad and Shahaboddin had conducted an experiment on the twitter data by gathering the tweets related to the political reviews based on these reviews the sentiment analysis is done to the tweets by using the Naive Bayes algorithm and SVM and had also provided a comparison of these two techniques [10]. Imane, Rochdi, Alexis and Abdessamad has gathered a tweet related to the US election which was held in 2016 in a large number of volume and stored in the HDFS and by building dictionaries and the classification of the data is been done, data is been pre-processed and the sentiment analysis is carried out to the data by using the Naïve Bayes algorithm [11]. Ankita and Anand conducted a sentiment analysis on the review provided by the customers during their flight gathered an US airline tweets from dataset called Kaggle dataset for six major airlines and the sentiment analysis are done by using seven different machine learning classification [12].…”

Section: Related Workmentioning

confidence: 99%

Dynamic Data Analysis and Decision Making on Twitter Data

Kumari*¹,

Kandukuri²

2019

IJITEE

View full text Add to dashboard Cite

It became a tedious task for the data analysts to make decisions on social networks. The existing approaches are not adequate to perform data pre-processing, analysis and decision making on the data dynamically. Therefore, this research aims to propose an approach to data analysis and decision making. The proposed approach emphasis on extracting tweets form twitter API (Application Program Interface), pre-processing the tweets by following seven pre-processing steps. The processed tweets are trained by NLTK (Natural Language Toolkit) and Text Blob are given to the sentiment analysis. Classification is done using the Naive Bayes algorithm to make a decision on processed tweets. The tweets which are related to “MeToo Movement” are considered primarily for decision making and satisfactory results are obtained. It is been observed that the proposed approach is accurate when compared to other approaches.

show abstract

“…big data, consumer confidence index, microblog, social media, user dictionary 1 | INTRODUCTION Psychological research has found that when individuals make cognitive evaluation, their emotions would be affected; at the same time, the emotion of individuals would in return influence the cognitive evaluation process. The sentiment analysis based on microblog texts mining is of great value to study users' satisfaction with commodities, attitudes towards social events, and individual psychological behavior characteristics (Bai et al, 2013;El Alaoui et al, 2018;He, 2013;Kadam et al, 2018;Li & Wu, 2010;Shayaa et al, 2018;Tausczik & Pennebaker, 2010). Based on the theory that emotions influence financial decisions, plenty of studies have combined social emotions with stock market trends (Koy & Akkaya, 2017).…”

Section: Introductionmentioning

confidence: 99%

Building consumer confidence index based on social media big data

Wang

Zhan

et al. 2019

Human Behav and Emerg Tech

View full text Add to dashboard Cite

This study takes microblog texts as the original data, builds Consumer Confidence Index (CCI) user dictionary, and calculates the CCI through different experimental conditions of the target words' frequency analysis. We explored the better way between “exact matching” and “fuzzy matching,” and tested the necessity of degree words and negative words. The main findings are as follows: (a) The numbers of principal components extracted in the six different matching methods and weight settings were similar, which were nine or 10 principal components, and the explanatory variance of composition was above 96%; (b) Comprehensively considered, the principle of “fuzzy matching target word + degree word” was the best calculation method; (c) When the validation set data to predict the CCI in next month which based on the optimal method was used, the difference between the index estimate and the official value was within 10%. To a certain degree, CCI based on microblog can be on behalf of consumers' subjective feelings as well as attitudes towards China's current and future economic situations.

show abstract

A novel adaptable approach for sentiment analysis on big social data

Cited by 119 publications

References 26 publications

Aspect Based Sentiments from Tweets using Co-Ranking Multi-Modal Natural Language Processing Methodologies

Aspect Based Sentiments from Tweets using Co-Ranking Multi-Modal Natural Language Processing Methodologies

Dynamic Data Analysis and Decision Making on Twitter Data

Building consumer confidence index based on social media big data

Contact Info

Product

Resources

About