Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

 
Session Overview
Session
B2: Tracking and Data Collection
Time:
Thursday, 16/Mar/2017:
10:45 - 11:45

Session Chair: Stefan Niebrügge, INNOFACT AG, Germany
Location: A 026

Show help for 'Increase or decrease the abstract text size'
Presentations

Deep Learning - Manage Online communication in the Age of Trolls

Hervé Flutto, Maurice Gonzenbach, Pascal de Buren

100 Celsius, Switzerland

Relevance & Research Question

"Something has changed – as globalization has marched on, [political] debate is taking place in a completely new media environment. Opinions aren't formed the way they were 25 years ago" as Angela Merkel stated in her speech to Parliament on November 23rd 2016.[1]

We are in a world where social media and peer-to-peer communication is taking the lead over curated contents. The exponential growth of trolls, fake news and chat bots raises the question how state-of-art Natural Language Processing powered by deep learning algorithms can be applied to curating online contents.

Methods & Data

We build on a wide array of self-developed as well as open source software, originating from research performed at ETH, Zurich.[2] Thereby we make use of state of the art machine learning technologies such as word embedding, convolutional or recurrent neural networks, but also more traditional, heuristic approaches.

We pre-train our model using publicly available German text data and fine tune it with a dataset obtained from a major Swiss-German online-newspaper.

Results

We have been able to establish leading sentiment analysis models on challenging data inputs such as Tweets, winning the 2016 Semeval competition (task 4a).[3] Building on those proven methods, we have since tackled the problem of detecting unwanted user comments in German language online newspapers. In particular, our system aims at detecting “trolls” who try to capture online discussions and manipulate the public opinion.

Added Value

This approach can be used to manage online communication through cost-effective and repeatable tools to complement human judgment on the quality and validity of online data. It can further be used to monitor and act upon customer satisfaction and sentiment at company or product level.

[1] http://www.usnews.com/news/features/news-video?ndn.trackingGroup=90080&ndn.siteSection=ndn1_usnews&ndn.videoId=31671176

[2] http://e-collecHon.ethbib.ethz.ch/show?type=dipl&nr=968

[3] https://www.inf.ethz.ch/news-and-events/spotlights/semeval2016.html

https://aclweb.org/anthology/S/S16/S16-1173.pdf


Flutto-Deep Learning - Manage Online communication in the Age-247.pdf

Mapping the Field of Automated Data Collection in the Web. Data Types, Collection Approaches and their Research Logic.

Jakob Jünger

University of Greifswald, Germany

Relevance & Research Question: Online communication makes the interaction of individuals, organizations and companies visible – because this interaction leaves data trails, or even consists of data itself. It is no surprise, therefore, that social scientists also work intensively on collecting online data. How the techniques used can be methodologically and epistemologically localized, however, is still unclear. This uncertainty is also reflected in the variety of terminology proposals. Concepts such as Computational Social Science (Lazer et al., 2009), Web Mining (Thelwall, 2009), or Digital Methods (Rogers, 2010) come into play. Furthermore, data collection methodology forms a diverse landscape regarding different types of data, collection methods and data providers (Keyling/Jünger 2016). The paper asks which methodological challenges as well as opportunities result from different types of data and collection methods.

Methods & Data: Based on experience when collecting online data in the field of political communication research three different approaches for automated data collection are discussed and backed up with examples: raw data, application programming interfaces and user interfaces. Each of these approaches is analyzed in terms of seven methodological dimensions: research object, analysis perpective, data level, abstraction, reactivity, structuring and availability.

Results: The analysis results in a classification scheme which helps with identifying specific methodological opportunities and challenges. Comparing different approaches makes clear that the data never speak for themselves. However, there seems to be a lack of standards with regard to, e.g. reliability and validity of the database or the description of the procedure, which sometimes seems to be ignored with references to numerically large datasets. In constrast dealing with smaller datasets may be more valuable under certain conditions.

Added Value: The paper adds value to the ongoing discussion by systematically mapping the landscape of automated data collection methods in the web. It brings to mind the necessity of dealing with quality criteria in Computational Social Science.


Jünger-Mapping the Field of Automated Data Collection in the Web Data Types, Collection Approaches and their .pdf

Wearable Research Technology: Tracking Tools for All Occasions?

Fabiola Gattringer, Manuela Schmid, Barbara Stiglbauer, Bernad Batinic

Johannes Kepler University Linz, Austria

Relevance & Research Question: Wearables and tracking tools are on the rise on the market, and prominent manufacturers introduce an improved device at least each year. Devices come in manifold forms, colors, and with different functionalities for virtually every occasion: from smart clothing over fashionable accessories to data glasses, with the capability to measure all kinds of scientifically interesting data. What possibilities offer these wearables and self-tracking technologies for scientific research, especially in the field of psychology and social sciences? More specifically, what added value has the collection of such objective biometric data in connection with subjective self-reported data via online surveys?

Methods & Data: We used a mixed methods approach: 1) An extensive literature research on studies with and about wearables and tools to measure physical and biometric data was conducted to provide a base for our literature review on possibilities for future scientific research. 2) A quantitative survey study (n = 98) was conducted to yield a first insight in the usability and acceptance of wearables in everyday life, especially working life.

Results: There are manifold devices on the market, and some have already been successfully used for research purposes in various scientific fields. In our review, we evaluate the pros and cons of these tools for scientific research in social sciences, and highlight promising research trends, as well as provide guidelines for how to avoid obstacles and achieve valid study results.

Added Value: This is, to our knowledge, the first attempt to provide an overview over the vast possibilities of wearable and self-tracking technology for scientific research purposes in social sciences, with a specific focus on previously conducted studies with such devices and an outlook on the usability for future, especially psychological, research. The review outlines possibilities and obstacles according to previously conducted research and additionally offers guidelines and recommendations for conducting future research endeavors.


Gattringer-Wearable Research Technology-196.pptx


 
Contact and Legal Notice · Contact Address:
Conference: GOR 17
Conference Software - ConfTool Pro 2.6.96
© 2001 - 2016 by H. Weinreich, Hamburg, Germany