Zengyi Shi

Bio: Zengyi Shi is an academic researcher from Beijing University of Posts and Telecommunications. The author has contributed to research in topics: Video production & Emotion classification. The author has an hindex of 1, co-authored 1 publications receiving 4 citations.

Improved Danmaku Emotion Analysis and Its Application Based on Bi-LSTM Model

Abstract: With the rapid development of social media, danmaku video provides a platform for users to communicate online. To some extent, danmaku video provides emotional timing information and an innovative method to analyze video data. In the age of big data, studying the characteristics of danmaku and its emotional tendencies can not only help us understand the psychological characteristics of users but also feedback the effective information of users to video platforms, which can help the platforms optimize related short video recommendations so that it can provide a more accurate solution for the selection of audiences during video production. However, danmaku is different from traditional comments. Current emotion classification methods are only suitable for two-dimensional classification which are not suitable for danmaku emotion analysis. Aiming at the problems such as the colloquialism, diversity, spelling errors, structural non-linearity informal language on the Internet, diversity of social topics, and context dependency of emotion analysis of the danmaku data, this paper proposes an improved emotion analysis model based on Bi-LSTM model to classify the further four-dimensional emotions of Pleasure, Anger, Sorrow and Joy. Furthermore, we add tags such as comment time and user name to the danmaku information. Experimental results show that the improved model has higher Accuracy, Recall, Precision, and F1-Score under the same conditions compared with the CNN and SVM. The classification effect of improved model is close to the SOTA. Experimental results also show that the improved model can be effectively applied to the analysis of irregular danmaku emotion.

Enhancing Semantic Relation Classification With Shortest Dependency Path Reasoning

Abstract: Relation Classification (RC) is a basic and essential task of Natural Language Processing. Existing RC methods can be classified into two categories: sequence-based methods and dependency-based methods. Sequence-based methods identify the target relation based on the overall semantics of the whole sentence, which will inevitably introduce noisy features. Dependency-based methods extract indicative word-level features from the Shortest Dependency Path (SDP) between given entities and attempt to establish a statistical association between the words and the target relations. This pattern relatively eliminates the influence of noisy features and achieves a robust performance on long sentences. Nevertheless, we observe that majority of relation classification processes involve complex semantic reasoning which is hard to be achieved based on the word-level statistical association. To solve this problem, we categorize all relations into atomic relations and composed-relations. The atomic relations are the basic relations that can be identified based on the word-level features, while the composed-relation requires to be deducted from multiple atomic relations. Correspondingly, we propose the Atomic Relation Encoding and Reasoning Model (ATERM). In the atomic relation encoding stage, ATERM groups the word-level features and encodes multiple atomic relations in parallel. In the atomic relation reasoning stage, ATERM establishes the atomic relation chain where relation-level features are extracted to identify composed-relations. Experiments show that our method achieves state-of-the-art results on the three most popular relation classification datasets – TACRED, TACRED-Revisit, and SemEval 2010 task 8 with significant improvements.

USee: An Online-Offline Hybird Danmaku Social System

Abstract: This paper presents a notion of location-sensitive online-offline hybrid social system What makes it distinct from other mobile social media is that it supports not simply online interaction but also online-offline hybrid interaction, and location sensitivity is a feature designed to lower cost for users to participant in offline activities We designed and implemented a prototype mobile Application called USee, and danmaku - an emerging socio-digital paradigm, was employed as the main user interface We introduce the design and implementation of USee, and conduct a preliminary study to understand how it works in reality Findings of the preliminary study suggest promises of the notion of location-sensitive online-offline hybrid social system We highlight that USee provides users a free and safe social place, in which users express high and unique enthusiasm and engagement This paper contributes to provide meaningful insights and design implication to mobile social media

Affective video recommender systems: A survey

Abstract: Traditional video recommendation provides the viewers with customized media content according to their historical records (e.g., ratings, reviews). However, such systems tend to generate terrible results if the data is insufficient, which leads to a cold-start problem. An affective video recommender system (AVRS) is a multidiscipline and multimodal human-robot interaction (HRI) system, and it incorporates physical, physiological, neuroscience, and computer science subjects and multimedia resources, including text, audio, and video. As a promising research domain, AVRS employs advanced affective analysis technologies in video resources; therefore, it can solve the cold-start problem. In AVRS, the viewers’ emotional responses can be obtained from various techniques, including physical signals (e.g., facial expression, gestures, and speech) and internal signals (e.g., physiological signals). The changes in these signals can be detected when the viewers face specific situations. The physiological signals are a response to central and autonomic nervous systems and are mostly involuntarily activated, which cannot be easily controlled. Therefore, it is suitable for reliable emotion analysis. The physical signals can be recorded by a webcam or recorder. In contrast, the physiological signals can be collected by various equipment, e.g., psychophysiological heart rate (HR) signals calculated by echocardiogram (ECG), electro-dermal activity (EDA), and brain activity (GA) from electroencephalography (EEG) signals, skin conductance response (SCR) by a galvanic skin response (GSR), and photoplethysmography (PPG) estimating users’ pulse. This survey aims to provide a comprehensive overview of the AVRS domain. To analyze the recent efforts in the field of affective video recommendation, we collected 92 relevant published articles from Google Scholar and summarized the articles and their key findings. In this survey, we feature these articles concerning AVRS from different perspectives, including various traditional recommendation algorithms and advanced deep learning-based algorithms, the commonly used affective video recommendation databases, audience response categories, and evaluation methods. Finally, we conclude the challenge of AVRS and provide the potential future research directions.

Live Comments Emotional Analysis based on EE-RNN

Abstract: Live comments, also known as video Danmaku, is a technique through which audiences can express their real-time feelings and opinions with rich emotional information. Emotional analysis results of live comments can more truly reflect the overall characteristics of the video, while user's feedback can be further exploited by extensive applications. Most of the existing live comments emotion classification methods do not fully consider either the real fine granularity or the explicit emotional knowledge of the on-screen comments text. Besides, existing machine learning methods and deep learning methods such as Long Short-Term Memory neural network and Convolutional Neural Network based models do not make full use of the semantic layer representation and emotional features of the text. In this paper, Enhanced ERNIE Deep Recurrent Neural Networks model (EE-RNN) is employed to complete the five-dimensional live comments emotional analysis. The model first obtains the general semantic embedding of the text through ERNIE and introduces external emotional knowledge to further enhance the semantic coding representation, and then uses improved RNN structure as well as attention mechanism to get an emotional enhanced high-level semantic feature representation. Experimental results on the live comments emotional classification dataset and NLPCC2014 emotional classification dataset show that the proposed model greatly improves the classification performance compared with the existing methods and can be used in real applications.

The impact of barrage system fluctuation on user interaction in digital video platforms: a perspective from signaling theory and social impact theory

Abstract: PurposeIn the present world of constant connectivity, the barrage system, as a system of real-time dynamic comments coupled with video content, has become a popular interactive system technology for video sharing platforms. This study investigates how barrage system fluctuation characteristics, namely, barrage fluctuation amplitude and frequency, impact user interaction.Design/methodology/approachThe research model was estimated with a fixed-effects regression applied to a longitudinal panel dataset collected from one of the most popular video sharing platforms in China (Bilibili.com).FindingsBarrage fluctuation frequency has positive effects on users' real-time (synchronous) barrage interaction and the traditional (asynchronous) comment interaction. Barrage fluctuation amplitude has a positive effect on users' real-time (synchronous) barrage interaction but a negative effect on traditional (asynchronous) comment interaction. In addition, the interaction effects of the barrage fluctuation frequency and the barrage fluctuation amplitude on user interaction show adverse effects.Originality/valueThe results revealed the impact of different barrage fluctuation characteristics on different forms of interaction and provide important theoretical contributions and managerial implications in terms of user interaction on video sharing platforms.

Albert-based sentiment analysis of movie review

Abstract: Movie reviews include the real evaluation of the movie by the public. Through these reviews, the audience can better judge whether the movie is worth watching. However, as the amount of data on movie reviews continues to grow, it takes a lot of manpower and material resources to manually analyze the emotional tendency of each movie review. As an important research field of machine learning, sentiment analysis focuses on extracting topic information from text reviews. The field of sentiment analysis is closely related to natural language processing and text mining. It can be successfully used to determine the reviewer's attitude towards various topics or the overall polarity of the review. As far as movie reviews are concerned, in addition to scoring movies digitally, they can also quantitatively enlighten us on the advantages and disadvantages of watching movies. This article uses the Albert model to build a classifier, and uses the "movie review dataset" issued by Stanford University for network training. Experiments show that the trained Albert model can reach an accuracy of 89.05% when performing sentiment analysis of movie reviews. Compared with the traditional LSTM and GRU, the accuracy of the Albert model is improved by 3%.