Leveraging sentiment analysis at the aspects level to predict ratings of reviews

Qiu, Junyi; Liu, Chuanhui; Li, Yinghong; Lin, Zhangxi

doi:10.1016/j.ins.2018.04.009

Cited by 46 publications

(15 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A peer validated the rating by consensus with the reviewer and it ranged through positive, slightly positive, neutral, slightly negative, and finally negative on a five-point Likert-type scale. This approach was adapted from Qiu et al (2018), who used a similar technique to categorize articles based on the sentiment expressed in the conclusion. In this context, the author assigned the sentiment rating based on how the flipped classroom pedagogy was perceived by the university students in the 22 articles.…”

Section: The Polarizing Effect Of Flipped Classroomsmentioning

confidence: 99%

Innovative Teaching: A Qualitative Review of Flipped Classrooms

Fuchs

2021

IJLTER

View full text Add to dashboard Cite

The extent and importance of web-based learning in higher education have increased tremendously in the last decade, triggered by new educational technologies and pedagogical approaches. Higher education lecturing has traditionally followed a teacher-centered approach, with lecturers giving classes in the classroom and students performing out-of-class activities. Under this traditional approach, the main actor in the teaching-learning process is the lecturer, while students play a passive role. The flipped classroom emphasizes the role of the student in the center of the learning environment and facilities an active learning pedagogy. This paper reviews the most recent case studies related to the flipped classroom approach in order to provide educators guidelines on the best practices, traits, and merits of the flipped classroom. A total of 22 case studies were included in this qualitative review of the flipped classroom. The methodological inquiry followed the PRISMA flow diagram that identified an initial pool of 3,764 articles. Upon identification of relevant case studies (n=22), a five-point Likert-type sentiment rating was assigned as the basis to structure the discussion. The rating was based on the students’ perceptions of the flipped classrooms as investigated in the articles. The assessment confirms that the majority of students have a positive perception of this learning pedagogy. However, there are concerns about increased workload for students, ambiguity in expected learning outcomes, and an initial barrier to actively contribute; these are the primary implications of this review.

show abstract

Section: The Polarizing Effect Of Flipped Classroomsmentioning

confidence: 99%

Innovative Teaching: A Qualitative Review of Flipped Classrooms

Fuchs

2021

IJLTER

View full text Add to dashboard Cite

show abstract

“…S. Akhtar et.al, focused on the stock salience and the asymmetric market influence of consumers' sentiments [1]. They have included asymmetric announcement influences of various customers' sentiments.…”

Section: Related Workmentioning

confidence: 99%

“…According to a recent report [1], macroeconomic news is capable to explain one-third of the variance in stock returns. Traditional classification models use the financial news to restrict the short-term predictive capability on future stock prices.…”

Section: Introductionmentioning

confidence: 99%

A New sentiment score based improved Bayesian networks for real-time intraday stock trend classification

Maguluri¹

2019

IJATCSE

View full text Add to dashboard Cite

The stock sentiment score is an important trend indicator for an enterprise and several factors may affect its value. Various events such as stock news, quarterly results and company announcements may differ in public emotion and sentiments, which may affect stock market trend. Due to variation in stock trend and news, stock market forecasting remains a challenging task in real-time market. However, most of the traditional sentiment analysis approaches are not directly predicting the sentiment score and its trend on the online stock market sites due to data sparsity and uncertainty problems. In order to overcome these problems, a hybrid probabilistic based stock sentiment prediction model is designed and implemented based on the real-time market data. In the proposed model, real-time stock data such as stocks technical, stocks news, stock momentum etc. are integrated to predict the buying or selling behavior of the investor. A novel stock sentiment score and data transformation methods are implemented to normalize the market trend for intraday data. Finally, an improved Bayesian networks technique is implemented on the filtered data to predict and test the trend of the stock on the filtered data. Experimental results show that the proposed framework is more efficient than the traditional sentiment classification models in terms of accuracy and error rate are concerned.

show abstract

“…In recent years, with the maturation of speech recognition technology, its application fields have become extensive and include common speech processing tasks such as machine translation, (1) question answering, (2) and sentiment analysis. (3) The related technologies of speech processing include machine learning, data mining, knowledge acquisition related to language processing, and linguistic research related to language computing. However, most existing speech recognition (4,5) and natural language processing (2,6) systems do not involve the control of industrial equipment.…”

Section: Introductionmentioning

confidence: 99%

Improved Rapid Automatic Keyword Extraction for Voice-based Mechanical Arm Control

Shih¹,

Lin²,

Jeng³

2021

Sensors and Materials

View full text Add to dashboard Cite

As smart voices gradually enter the human living environment, people can interact with controlled systems through simple commands without the need for a screen. Even people without specialist training can use voice sensing control to easily complete the operation of a robot arm while avoiding incorrect operation caused by insufficient professional knowledge. In this study, we use the Google speech recognition engine to convert input speech into text information, and then perform speech recognition, word segmentation, and part-of-speech (POS) tagging. We also propose an improved rapid automatic keyword extraction (IRAKE) method, which uses the word string matching feature in the dictionary method to correspond to the relevant execution action function. The experimental results show that the accuracy rate of the proposed IRAKE method is 16% higher than that of the traditional rapid automatic keyword extraction (RAKE) method.

show abstract

Leveraging sentiment analysis at the aspects level to predict ratings of reviews

Cited by 46 publications

References 15 publications

Innovative Teaching: A Qualitative Review of Flipped Classrooms

Innovative Teaching: A Qualitative Review of Flipped Classrooms

A New sentiment score based improved Bayesian networks for real-time intraday stock trend classification

Improved Rapid Automatic Keyword Extraction for Voice-based Mechanical Arm Control

Contact Info

Product

Resources

About