No abstract

Background: Assisting moderators to triage harmful posts in Internet Support Groups is relevant to ensure its safe use. Automated text classification methods analysing the language expressed in posts of online forums is a promising solution.Methods: Natural Language Processing and Machine Learning technologies were used to build a triage post classifier using a dataset from Reachout.com mental health forum for young people.Results: When comparing with the state-of-the-art, a solution mainly based on features from lexical resources, received the best classification performance for the crisis posts (52%), which is the most severe class. Six salient linguistic characteristics were found when analysing the crisis post; (1) posts expressing hopelessness, (2) short posts expressing concise negative emotional responses, (3) long posts expressing variations of emotions, (4) posts expressing dissatisfaction with available health services, (5) posts utilising storytelling, and (6) posts expressing users seeking advice from peers during a crisis. Conclusion:It is possible to build a competitive triage classifier using features derived only from the textual content of the post. Further research needs to be done in order to translate our quantitative and qualitative findings into features, as it may improve overall performance.

Due to the high maintenance cost of traditional online auction system, poor load balance ability and down problems occurring in the peak value of system access, this paper aims to propose a solution to transit from traditional online auction system to cloud computing to get higher work efficiency, lower expenditure and less energy cost. The GAE platform is used to deploy application of the overall framework of online auction system in cloud computing, including development environment as well as online auction system components, and software architecture. The online auction negotiation algorithms in cloud computing are also proposed. Based on these key technologies, the business processes of the online auction system in cloud computing is designed, including users' login system, starting an auction, bidding processes, online auction data storage, and logout system. The online auction system constructed on GAE platform with cloud computing resources and storage ability can reduce the pressure of terminal equipment, which is more robust than traditional online auction system facing users' changeable needs in the process of online auction.

Fake online reviews are so prevalent that e-commerce platforms attempt to control it from affecting the trustworthiness between buyers and sellers. The issue has also attracted sporadic scholarly endeavor to understand this new field. To address this issue, we propose a new model to examine three interrelated stakeholders of e-Commerce platforms: experienced buyers, future buyers and the online sellers in terms of purchasing behaviors and sales with three objectives. Experienced buyers influence future consumers’ behaviors and increase sales from sellers. Using data collected from the largest online e-commerce platform in China, we test relevant hypotheses. Our findings show that experienced buyers and their positive reviews increase future buyers’ purchasing and promote corporate sales. These findings contribute knowledge to the online feedback mechanism and literature on fake review studies. This study also provides a novel method to help buyers avoid fake online review from a market structure perspective.

Background: Assisting moderators to triage harmful posts in Internet Support Groups is relevant to ensure its safe use. Automated text classification methods analysing the language expressed in posts of online forums is a promising solution. Methods: Natural Language Processing and Machine Learning technologies were used to build a triage post classifier using a dataset from Reachout.com mental health forum for young people. Results: When comparing with the state-of-the-art, a solution mainly based on features from lexical resources, received the best classification performance for the crisis posts (52%), which is the most severe class. Six salient linguistic characteristics were found when analysing the crisis post; 1) posts expressing hopelessness, 2) short posts expressing concise negative emotional responses, 3) long posts expressing variations of emotions, 4) posts expressing dissatisfaction with available health services, 5) posts utilising story-