Filtering by
- All Subjects: Cyberbullying
- All Subjects: Semantic computing
- Creators: Tong, Hanghang
Background: Cyberbullying and cyber-victimization are rising problems and are associated with increased risk for mental health problems in children. Methods for addressing cyberbullying are limited, however, interventions focused on promoting appropriate parental mediation strategies are a promising solution supported by evidence and by guided by the Theory of Parenting Styles.
Objective: To provide an educational session to parents of middle school students that promotes effective methods of preventing and addressing cyberbullying incidents. Design: The educational sessions were provided to eight parents middle school student. Surveys to assess parent perception of and planned response to cyberbullying incidents and Parent Adolescent Communication Scale (PACS) scores were collected pre-presentation, post-presentation, and at one-month follow up.
Results: Data analysis of pre- and post-presentation PACS using a Wilcoxon test found no significant difference (Z = -.405, p >.05). There was not enough response to the 1-month follow-up to perform a data analysis on follow-up data.
Conclusions: Due to low attendance and participation in the follow-up survey the results of this project are limited. However, parents did appear to benefit from communicating concerns about cyberbullying with school officials. Future studies should examine if a school-wide anti-cyberbullying program that actively involves parents effects parental response to cyberbullying.
Many video feature extraction algorithms have been purposed, such as STIP, HOG3D, and Dense Trajectories. These algorithms are often referred to as “handcrafted” features as they were deliberately designed based on some reasonable considerations. However, these algorithms may fail when dealing with high-level tasks or complex scene videos. Due to the success of using deep convolution neural networks (CNNs) to extract global representations for static images, researchers have been using similar techniques to tackle video contents. Typical techniques first extract spatial features by processing raw images using deep convolution architectures designed for static image classifications. Then simple average, concatenation or classifier-based fusion/pooling methods are applied to the extracted features. I argue that features extracted in such ways do not acquire enough representative information since videos, unlike images, should be characterized as a temporal sequence of semantically coherent visual contents and thus need to be represented in a manner considering both semantic and spatio-temporal information.
In this thesis, I propose a novel architecture to learn semantic spatio-temporal embedding for videos to support high-level video analysis. The proposed method encodes video spatial and temporal information separately by employing a deep architecture consisting of two channels of convolutional neural networks (capturing appearance and local motion) followed by their corresponding Fully Connected Gated Recurrent Unit (FC-GRU) encoders for capturing longer-term temporal structure of the CNN features. The resultant spatio-temporal representation (a vector) is used to learn a mapping via a Fully Connected Multilayer Perceptron (FC-MLP) to the word2vec semantic embedding space, leading to a semantic interpretation of the video vector that supports high-level analysis. I evaluate the usefulness and effectiveness of this new video representation by conducting experiments on action recognition, zero-shot video classification, and semantic video retrieval (word-to-video) retrieval, using the UCF101 action recognition dataset.