Manuela Pollak, Gabriele Anderst-Kotsis,
"E-Mail Monitoring and Management with MS Social Bots"
, in Maria Indrawan-Santiago, Ivan Luiz Salvadori, Matthias Steinbauer, Ismail Khalil, Gabriele Anderst-Kotsis (Eds.): Proceedings of the 19th International Conference on Information Integration and Web-based Applications & Services (iiWAS 2017), ACM, Seite(n) 234-240, 2017, ISBN: 978-1-4503-5299-4
E-Mail Monitoring and Management with MS Social Bots
Sprache des Titels:
Proceedings of the 19th International Conference on Information Integration and Web-based Applications & Services (iiWAS 2017)
Social Bots are software robots implementing algorithms that autonomously produce content and interact with users of social networks or mimic the behavior of humans in multiple other forms of digital communication. If they act on behalf of a specific user of party in the communication, this user is sometimes called the "owner" of the bot. Social bots are typically designed to be, benevolent and even useful but some of them are built to manipulate and deceive social users.
Since spring 2016 Microsoft provides the technology Microsoft Bot Framework. With this framework developers can create own social bot applications easily in Visual Studio with the language C#. The bot application can be connected to various social channels like Facebook, Twitter, Microsoft Groups or an Office 365 accounts.
In this paper we demonstrate how the Microsoft technology can be combined with methods from artificial intelligence to become even more productive and useful for the owner of the bot. The main purpose of the implemented bot application of "Email Monitoring and Management with Microsoft Social Bot" is filtering appointment requests out of incoming Emails and to analyse them with AI. After the decision-making process succeeded, the bot creates a new calendar event in the Office 365 calendar and answers automatically to the appointment requests sender. For the decision-making process we compare the methods Decision Tree with ID3 and Naïve Bayes from the classification of the AI is implemented. The evaluation with measurement of the sample data shows that the Decision Tree is quite effective an even in the performance the Decision Tree is faster during the decision-making process.
The focus of the paper is on proofing the technical feasibility and implementation of "intelligent" social bots with state of the art technology.