{"title":"How to select information that matters: a comparative study on active learning strategies for classification","authors":"C. Beyer, G. Krempl, V. Lemaire","doi":"10.1145/2809563.2809594","DOIUrl":"https://doi.org/10.1145/2809563.2809594","url":null,"abstract":"Facing ever increasing volumes of data but limited human annotation capabilities, active learning strategies for selecting the most informative labels gain in importance. However, the choice of an appropriate active learning strategy itself is a complex task that requires to consider different criteria such as the informativeness of the selected labels, the versatility with respect to classification algorithms, or the processing speed. This raises the question, which combinations of active learning strategies and classification algorithms are the most promising to apply. A general answer to this question, without application-specific, label-intensive experiments on each dataset, is highly desirable, as active learning is applied in situations with limited labelled data. Therefore, this paper studies several combinations of different active learning strategies and classification algorithms and evaluates them in a series of comparative experiments.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"439 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75177455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yordan Terziev, Marian Benner-Wickner, Tobias Brückmann, V. Gruhn
{"title":"Ontology-based recommender system for information support in knowledge-intensive processes","authors":"Yordan Terziev, Marian Benner-Wickner, Tobias Brückmann, V. Gruhn","doi":"10.1145/2809563.2809600","DOIUrl":"https://doi.org/10.1145/2809563.2809600","url":null,"abstract":"Knowledge-intensive processes are difficult to support because of their complexity, high variability and unpredictable information requirements. Therefore such process types are handled manually by knowledge workers with expertise in the domain. Yet to make informed decisions, knowledge workers require a multitude of domain specific, case-related information. This often leads to a time-consuming search for information and knowledge required to address the issues occurring in the case. To reduce the time spent searching for information, we propose an ontology-based recommender system that provides case-related information based on documents gathered in accumulated similar cases. The recommender system builds models of domain specific concepts for past cases as well as for the current case, which are used for case similarity calculation. To evaluate the performance of parts of our approach we used the OHSUMED document collection and compared the cosine similarity measure of ontological case model against textual case model.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"67 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90367450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Digital and open innovation: implications for business models and research strategies of private and public organizations","authors":"K. Leitner","doi":"10.1145/2809563.2809609","DOIUrl":"https://doi.org/10.1145/2809563.2809609","url":null,"abstract":"While new ways of doing research and innovation such as open science, Science 2.0, user innovation or crowdsourcing have been intensively discussed in the past few years, little systematic analysis and exploration of their wider positive, but also problematic effects for business and science has been done. Based on the findings from two European foresight projects critical aspects of changing research and innovation models and their challenges for research management and policy are discussed. A key challenge for companies, universities and research organizations is to adopt of new business models and define new collaboration strategies.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88692884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Aggregated parallel coordinates: integrating hierarchical dimensions into parallel coordinates visualisations","authors":"K. Andrews, Majda Osmic, Gerhard Schagerl","doi":"10.1145/2809563.2809588","DOIUrl":"https://doi.org/10.1145/2809563.2809588","url":null,"abstract":"Aggregated Parallel Coordinates (APC) are an extension of standard parallel coordinates, which supports the visualisation and exploration of hierarchies within numerical dimensions. Such datasets can occur when data is available at several granularities and these can be grouped or aggregated in some way (mean, sum, max) to form higher levels of abstraction. While existing parallel coordinates techniques can be used to visualise individual dimensions of such data, they have no provision for interactively expanding and collapsing such hierarchically aggregated dimensions. In race car engineering, specialised simulation software is used to derive a car setup for optimal performance during a race. Such simulations generate complex, high-dimensional datasets with a large number of records, many of which are inherently hierarchically grouped. The APC software was implemented in C♯ as a Microsoft WPF user control library. It has been integrated into AVL's data visualisation tool called SimBook and is being used to explore the output of race car simulations.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"45 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79459989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"ICT and science 2.0: technology-mediated trends and characteristics of new scientific practices","authors":"Eric Prem","doi":"10.1145/2809563.2809590","DOIUrl":"https://doi.org/10.1145/2809563.2809590","url":null,"abstract":"ICT is currently changing scientific processes from data collection, data interpretation, formation of hypotheses and theories, testing and validation to the dissemination of research results. The relation of ICT and new scientific practices is not just accidental; Science 2.0 -- defined as 'open and digital science' inherits features and characteristics from today's ICT. In this paper, we discuss these features from the point of view of epistemology: which basic components of the scientific process are likely to undergo changes and where will the current trends lead Science 2.0?","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"45 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83233650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Collaborative writing in the context of science 2.0","authors":"Kaja Scheliga","doi":"10.1145/2809563.2809581","DOIUrl":"https://doi.org/10.1145/2809563.2809581","url":null,"abstract":"The increasing number of co-authored academic papers points to the importance of collaborative writing in contemporary research. Digital technologies add a new dimension to collaborative writing by providing co-authors with access to the same document and enabling co-authors to edit the shared text at the same time. The availability of web-based tools for collaborative writing prompts the question of the extent to which researchers incorporate these tools into their scholarly practices. Based on my statistical analysis of the data from the Science 2.0 Survey (2014), conducted in cooperation with the Leibniz Research Alliance Science 2.0, I examine the usage of digital technologies in the process of collaborative writing among researchers in Germany. I use the concepts of asynchronous and synchronous modes of writing, derived from the field of Computer Supported Cooperative Work, to discuss collaborative writing strategies in the context of Science 2.0. My study shows that researchers use a mixture of different writing strategies and that they tend to use the same tool for different writing strategies. Moreover, I discuss researchers' attitudes towards online text editors. In reflecting on collaborative writing, I consider both the technological and social aspects.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"46 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75030270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Influence of time on user profiling and recommending researchers in social media","authors":"Chifumi Nishioka, Gregor Große-Bölting, A. Scherp","doi":"10.1145/2809563.2809601","DOIUrl":"https://doi.org/10.1145/2809563.2809601","url":null,"abstract":"We conduct two experiments to compare different scoring functions for extracted user interests and measure the influence of using older data. We apply our experiments in the domains of computer science and medicine. The first experiment assesses similarity scores between a user's social media profile and a corresponding user's publication profile, in order to evaluate to which extend a user's social media profile reflects his or her professional interests. The second experiment recommends related researchers profiled by their publications based on a user's social media profile. The result revealed that while the functions using spreading activation produce large similarity scores between a user profile and publication profile, the scoring functions with statistical methods (e.g., an extension of BM25 with spreading activation) perform best for recommendation. In terms of the temporal influence, the older data have almost no influence on the performance in the medicine dataset. However, in the computer science dataset, while there is a positive influence in the first experiment, the second experiment demonstrated a negative influence when adding too old data.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74805406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Processing manufacturing knowledge with ontology-based annotations and cognitive architectures","authors":"Rebekka Alm, Mario Aehnelt, B. Urban","doi":"10.1145/2809563.2809576","DOIUrl":"https://doi.org/10.1145/2809563.2809576","url":null,"abstract":"Advanced manufacturing promises an evolution of industrial production processes. However, today's manufacturing systems lack a common strategy on how to combine factual, procedural, and conceptual knowledge in order to streamline production processes. This specifically applies for manufacturing assembly assistance where the major share of procedural and conceptual knowledge is not yet automatically processable. In our paper we propose the usage of ontology-based annotations as missing link between the tacit knowledge of the worker and the intelligent assistance system. We show the deeper integration of conceptual knowledge modeled in ontology-based annotations with procedural knowledge in cognitive architectures. Additionally, in our approach annotations act as a mean of communication between the workers and with the system. We show key aspects of a prototypical integration of our approach within a smart assembly assistance system which supplies the worker with task related information.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"110 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76087880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Business process oriented learning: a collaborative approach of organisational learning","authors":"R. Woitsch, Nesat Efendioglu","doi":"10.1145/2809563.2809578","DOIUrl":"https://doi.org/10.1145/2809563.2809578","url":null,"abstract":"Process Oriented Training and Learning can be applied in two different approaches: (a) processes describing the methodology of training and learning as well as (b) processes describing the organizational context that need to be learned. This paper intrudes the results of the EU project Learn PAd that developed prototypes of modelling tools enabling business processes for learning and training. Flexibility of business processes have been introduced with case management and knowledge artefacts of PROMOTE had been integrated to provide a complete modelling environment fulfilling the identified 101 requirements for the modelling language. The local deployment and the Web-based deployment of the developed prototypes are introduced and the development space that enables collaborative participation of the development and improvement of the prototypes on ADOxx.org is introduced.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"79 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83803201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Collaborative process maturing support by mining activity streams","authors":"Christian Ochsenkühn, R. Peinl","doi":"10.1145/2809563.2809583","DOIUrl":"https://doi.org/10.1145/2809563.2809583","url":null,"abstract":"Usually, knowledge workers are said to not benefit from business process management (BPM) systems, since their main tasks are weakly structured and not representable by a workflow. However, not all of their tasks are equally weak structured, and with adaptive case management (ACM) solutions, a new category of tools came up to support those processes, even if they are weakly structured. This paper introduces an approach to support the creation of cases for ACM engines by mining activities from an activity stream and suggesting tasks that a knowledge worker can use to create a case. Furthermore, the approach supports maturing of the case towards a workflow by detecting repeating sequences in the execution of tasks and suggesting sub processes for the case which is possible with the case management model and notation (CMMN) together with the business process model and notation (BPMN). To allow for further enhancement of the cases, the ACM solution is extended with social collaboration features, so that people working on the case can comment and rate single tasks. The goal is to show that it is possible to establish ties from social activities and Web 2.0 to ACM and BPM. The presented solution uses a graph database as a basis for activity mining.","PeriodicalId":20526,"journal":{"name":"Proceedings of the 15th International Conference on Knowledge Technologies and Data-driven Business","volume":"14 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2015-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82380206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}