Sentiment Analysis System for Roman Urdu

Mehmood, Khawar; Essam, Daryl; Shafi, Kamran

doi:10.1007/978-3-030-01174-1_3

Cited by 26 publications

(19 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another work is done in the domain of sentimental analysis which is not in English, Chinese or scientific domain but specifically in the Urdu domain [59], as very less work was done in the Urdu language [56]. They used the data set based on Urdu reviews related to movies, politics, mobile, dramas and miscellaneous domains extracted using scrapers as well as manual.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The decision is constructed by selecting the data from the data set randomly [3]. The information gain is calculated for all values and the feature with the highest information gain value becomes the tree's root [59] and the whole tree is constructed by finding the features for the next level again and again. The fast decision tree algorithm is developed by [38].…”

Section: G Classification Classifiersmentioning

confidence: 99%

“…Called lazy learner because its training phase contains nothing but storing all the training examples as classifiers [77]. KNN requires a lot of memory while storing the training values [59]. The performance issue of KNN can also be solved by efficient estimations of parameters [51].…”

Section: G Classification Classifiersmentioning

confidence: 99%

See 2 more Smart Citations

Scientific Text Sentiment Analysis using Machine Learning Techniques

Raza¹,

Faizan²,

Hamza³

et al. 2019

IJACSA

View full text Add to dashboard Cite

Over time, textual information on the World Wide Web (WWW) has increased exponentially, leading to potential research in the field of machine learning (ML) and natural language processing (NLP). Sentiment analysis of scientific domain articles is a very trendy and interesting topic nowadays. The main purpose of this research is to facilitate researchers to identify quality research papers based on their sentiment analysis. In this research, sentiment analysis of scientific articles using citation sentences is carried out using an existing constructed annotated corpus. This corpus is consisted of 8736 citation sentences. The noise was removed from data using different data normalization rules in order to clean the data corpus. To perform classification on this data set we developed a system in which six different machine learning algorithms including Naïve-Bayes (NB), Support Vector Machine (SVM), Logistic Regression (LR), Decision Tree (DT), K-Nearest Neighbor (KNN) and Random Forest (RF) are implemented. Then the accuracy of the system is evaluated using different evaluation metrics e.g. F-score and Accuracy score. To improve the system' accuracy additional features selection techniques, such as lemmatization, n-graming, tokenization, and stop word removal are applied and found that our system provided significant performance in every case compared to the base system. Our method achieved a maximum of about 9% improved results as compared to the base system.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: G Classification Classifiersmentioning

confidence: 99%

See 1 more Smart Citation

Scientific Text Sentiment Analysis using Machine Learning Techniques

Raza¹,

Faizan²,

Hamza³

et al. 2019

IJACSA

View full text Add to dashboard Cite

show abstract

“…While bag-of-words based feature representation approaches face the problem of data sparsity, randomly initialized word embeddings do not capture the semantics of text and fail to overshadow the performance of pre-trained neural word embeddings. Moreover, most of the existing work [19], [54], [20], [61], [55], [59], [56], [57] employs machine learning based methodologies, only Ghulam et al [62] have utilized a deep learning based approach for the task of Roman Urdu sentiment analysis. Also, no public benchmark dataset is available for Roman Urdu sentiment analysis…”

Section: Contributions: a Review Of Novel Work With Anticipated Imentioning

confidence: 99%

A Precisely Xtreme-Multi Channel Hybrid Approach for Roman Urdu Sentiment Analysis

et al. 2020

View full text Add to dashboard Cite

In order to accelerate the performance of various Natural Language Processing tasks for Roman Urdu, this paper for the very first time provides 3 neural word embeddings prepared using most widely used approaches namely Word2vec, FastText, and Glove. The integrity of generated neural word embeddings is evaluated using intrinsic and extrinsic evaluation approaches. Considering the lack of publicly available benchmark datasets, it provides a first-ever Roman Urdu public dataset which consists of 3241 sentiments annotated against positive, negative, and neutral classes. To provide benchmark baseline performance over the presented dataset for Roman Urdu sentiment analysis, we adapt diverse machine learning (Support Vector Machine, Logistic Regression, Naive Bayes), deep learning (convolutional neural network, recurrent neural network), and hybrid deep learning approaches. Performance impact of generated neural word embeddings based representation is compared with other most widely used bag of words based feature representation approaches using diverse machine and deep learning classifiers. In order to improve the performance of Roman Urdu sentiment analysis, it proposes a novel precisely extreme multi-channel hybrid methodology which makes use of convolutional and recurrent neural networks along with pre-trained neural word embeddings. The proposed hybrid approach outperforms adapted machine learning approaches by the significant figure of 9% and deep learning approaches by the figure of 4% in terms of F1-score. INDEX TERMS Fast-Text, Glove, Pretrain word embeddings for Roman Urdu, Roman Urdu Sentiment Analysis, Word2Vec,

show abstract

“…In scientific literature, citation sentences are often neutral in terms of opinion, either because they critically define an algorithm, strategy or technique, or because they favor a fact or argument [3]. [13] have worked on Sentiment Analysis of Roman Urdu. Most of the research works have been done on different subjects like "English", and "Chinese" etc.…”

Section: Literature Reviewmentioning

confidence: 99%

Scientific VS Non-Scientific Citation Annotational Complexity Analysis using Machine Learning Classifiers

Raza¹,

Faizan²,

Akhtar³

et al. 2020

IJACSA

View full text Add to dashboard Cite

This paper evaluates the citation sentences' annotation complexity of both scientific as well as non-scientific text related articles to find out major complexity reasons by performing sentiment analysis of scientific and non-scientific domain articles using our own developed corpora of these domains separately. For this research, we selected different data sources to prepare our corpora in order to perform sentimental analysis. After that, we have performed a manual annotation procedure to assign polarities using our defined annotation guidelines. We developed a classification system to check the quality of annotation work for both domains. From results, we have found that the scientific domain gave us more accurate results than the non-scientific domain. We have also explored the reasons for less accurate results and concluded that non-scientific text especially linguistics is of complex nature that leads to poor understanding and incorrect annotation.

show abstract

Sentiment Analysis System for Roman Urdu

Cited by 26 publications

References 22 publications

Scientific Text Sentiment Analysis using Machine Learning Techniques

Scientific Text Sentiment Analysis using Machine Learning Techniques

A Precisely Xtreme-Multi Channel Hybrid Approach for Roman Urdu Sentiment Analysis

Scientific VS Non-Scientific Citation Annotational Complexity Analysis using Machine Learning Classifiers

Contact Info

Product

Resources

About