Voice of urban park visitors: exploring destination attributes influencing behavioural intentions through online review mining

Only the label corresponding to the maximum value of the fully connected layer is used as the output category when a neural network performs classification tasks. When the maximum value of the fully connected layer is close to the sub-maximum value, the classification obtained by considering only the maximum value and ignoring the sub-maximum value is not completely accurate. To reduce the noise and improve classification accuracy, combining the principles of fuzzy reasoning, this paper integrates all the output results of the fully connected layer with the emotional tendency of the text based on the dictionary to establish a multi-modal fuzzy recognition emotion enhancement model. The provided model considers the enhancement effect of negative words, degree adverbs, exclamation marks, and question marks based on the smallest subtree on the emotion of emotional words, and defines the global emotional membership function of emojis based on the corpus. Through comparing the results of CNN, LSTM, BiLSTM and GRU on Weibo and Douyin, it is shown that the provided model can effectively improve the text emotion recognition when the neural network output result is not clear, especially for long texts.

Section: Multimodal Sentiment Analysismentioning

confidence: 99%

A multi-modal and multi-scale emotion-enhanced inference model based on fuzzy recognition

Qiu

Yan

2021

“…The Keras framework is used for keyword extraction and vectorization. Keras is an open source artificial neural network library written in Python that can be used as a highlevel application program interface for TensorFlow, Microsoft-CNTK and Theano to design, debug, evaluate, apply and visualize deep learning models [23][24][25]. Keywords can describe the theme or primary content of the text.…”

Section: Keyword Extraction and Vectorizationmentioning

confidence: 99%

Product-harm crisis intelligent warning system design based on fine-grained sentiment analysis of automobile complaints

Wei

Zhou

2021

Due to the huge losses caused by product-harm crises and subsequent recalls in the automobile industry, companies must urgently design a product-harm crisis warning system. However, the designs of existing warning systems use the recurrent neural network algorithm, which suffers from gradient disappearance and gradient explosion issues. To compensate for these defects, this study uses a long and short-term memory algorithm to achieve a final prediction accuracy of 90%. This study contributes to the research and design of automatic crisis warning systems by considering sentiment and improving the accuracy of automobile product-harm crisis prediction.

“…Due to the huge amount of text data from various sources, text mining techniques have been applied in several fields such as mining online reviews to discover the behavioral intentions of park visitors [16], mining online reviews for customer satisfaction analysis [5], online review classification [7], user-generated content analysis [31], service blueprint development [25] and supply chain risk management [6]. Moro et al [20] used text mining software, Tools for Innovation Monitoring, to recognize the evolving technologies of renewable energy.…”

Section: Introductionmentioning

confidence: 99%

Exploring technology opportunities and evolution of IoT-related logistics services with text mining

Chen

2021

Many IoT technologies have been applied in the logistics industry in recent years, and they have had a substantial impact on many sectors such as shipping, air freight, warehousing, inventory, etc. Exploring technology opportunities and carrying out technological trend analysis are essential for IoT’s evolution, and there are many techniques or methods for doing so. In this paper, data analysis and text mining techniques, technology opportunity analysis (TOA) and technology-service evolution analysis (TSEA) have been applied to analyze and observe IoT technologies’ and services’ evolution. Academic journals, market reports, and patents have been collected and reviewed on the topic of IoT in the logistics field in this paper. Moreover, by using TOA, technology opportunities have been analyzed to explore IoT-related logistics services. The results of TOA, for example, show that cloud technology is essential to develop smart logistics services, and communication RFID technologies are key to developing information logistics services. Finally, TSEA enables the observation of IoT technology and logistics service evolution by combining unstructured and semi-structured data from text documents. Observing the results of TSEA, the evolution of IoT in logistics is identified, and the results of TSEA also confirm those of TOA using unstructured or semi-structured text data from documents only. The results of this paper are discussed and compared with those of some previous review studies. In summary, the results of this paper provide methodological guidelines on this topic for a comprehensive understanding of IoT-related logistics services.