Coupling topic modelling in opinion mining for social media analysis

Zhou, Xujuan; Tao, Xiaohui; Rahman, Mostafijur; Zhang, Ji

doi:10.1145/3106426.3106459

Cited by 24 publications

(10 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It is because SVM works on the principle of structural risk minimization as compare to the Naive Bayes classifier which performs efficiently where features are highly dependent. K-mean algorithm is used where opinions are related to each other, and there is a need to cluster them, as done in paper Zhou et al (2017) describes how OM is combined with topic modeling. The paper Babu et al (2017) targets the problems related to noisy data, teaches how to make a cluster of related tweets, and handles unstructured data.…”

Section: Sum Of Word Vectors (Swv)mentioning

confidence: 99%

A novel category detection of social media reviews in the restaurant industry

et al. 2020

View full text Add to dashboard Cite

Social media platforms have enabled users to share their thoughts, ideas, and opinions on different subject matters and meanwhile generate lots of information which can be adopted to understand people's emotion towards certain products. This information can be effectively applied for Aspect Category Detection (ACD). Similarly, people's emotions and recommendation based Artificial Intelligence (AI) powered systems are in trend to assist vendors and other customers to improve their standards. These systems have applications in all sorts of business available on multiple platforms. However, the current conventional approaches fail in providing promising results. Thus, in this paper, we propose novel convolutional attention based bidirectional modified LSTM by combining the techniques of the next word, next sequence, and pattern prediction with ACD. The proposed approach extracts significant features from public reviews to detect entity and attributes pair, which are treated as a sequence or pattern from a given opinion. Next, we trained our word vectors with the proposed model to strengthen the ACD process. Empirically, we compare the approach with the state-of-theart ACD models that use SemEval-2015, SemEval-2016, and SentiHood datasets. Results show that the proposed approach effectively achieve 78.96% F1-Score on SemEval-2015, 79.10% F1-Score on SemEval-2016, and 79.03% F1-Score on Sen-tiHood which is higher than the existing approaches.

show abstract

Section: Sum Of Word Vectors (Swv)mentioning

confidence: 99%

A novel category detection of social media reviews in the restaurant industry

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Data mining is defined as a process that uses mathematical, statistical, artificial intelligence and machine learning techniques to extract and identify useful information and subsequently gain knowledge from databases. Data mining algorithms have been widely used in range of research fields such as healthcare and medicine [5][6][7], sentiment analysis [8,9], education [10] etc. The purpose of applying data mining in bank industry is to use the available data to retain its best customers and to identify opportunities sell them additional services.…”

Section: Introductionmentioning

confidence: 99%

A Case Study of Predicting Banking Customers Behaviour by Using Data Mining

Zhou

Bargshady

Abdar

et al. 2019

2019 6th International Conference on Behavioral, Economic and Socio-Cultural Computing (BESC)

Self Cite

View full text Add to dashboard Cite

Data Mining (DM) is a technique that examines information stored in large database or data warehouse and find the patterns or trends in the data that are not yet known or suspected. DM techniques have been applied to a variety of different domains including Customer Relationship Management CRM). In this research, a new Customer Knowledge Management (CKM) framework based on data mining is proposed. The proposed data mining framework in this study manages relationships between banking organizations and their customers. Two typical data mining techniques-Neural Network and Association Rules-are applied to predict the behavior of customers and to increase the decision-making processes for recalling valued customers in banking industries. The experiments on the real world dataset are conducted and the different metrics are used to evaluate the performances of the two data mining models. The results indicate that the Neural Network model achieves better accuracy but takes longer time to train the model.

show abstract

“…The obtained information extracted from raw data is a turning point in improving services to individuals. Data mining and machine learning algorithms have been widely used in range of research fields such as healthcare [1][2][3][4], medicine [5][6][7], economics [8,9], decision support systems [10,11], sentiment analysis [12][13][14] etc.…”

Section: Introductionmentioning

confidence: 99%

An Ensemble-Based Decision Tree Approach for Educational Data Mining

Abdar

Zomorodi‐Moghadam

Zhou

2018

2018 5th International Conference on Behavioral, Economic, and Socio-Cultural Computing (BESC)

Self Cite

View full text Add to dashboard Cite

Nowadays, data mining and machine learning techniques are applied to a variety of different topics (e. g., healthcare and disease, security, decision support, sentiment analysis, education, etc.). Educational data mining investigates the performance of students and gives solutions to enhance the quality of education. The aim of this study is to use different data mining and machine learning algorithms on actual data sets related to students. To this end, we apply two decision tree methods. The methods can create several simple and understandable rules . Moreover, the performance of a decision tree is optimized by using an ensemble technique named Rotation Forest algorithm. Our findings indicate that the Rotation Forest algorithm can enhance the performance of decision trees in terms of different metrics. In addition, we found that the size of tree generated by decision trees ensemble were bigger than simple ones. This means that the proposed methodology can reveal more information concerning simple rules.

show abstract

Coupling topic modelling in opinion mining for social media analysis

Cited by 24 publications

References 14 publications

A novel category detection of social media reviews in the restaurant industry

A novel category detection of social media reviews in the restaurant industry

A Case Study of Predicting Banking Customers Behaviour by Using Data Mining

An Ensemble-Based Decision Tree Approach for Educational Data Mining

Contact Info

Product

Resources

About