Data and Web Science

Ontology matching consists of finding correspondences between ontology entities. OAEI campaigns aim at comparing ontology matching systems on precisely defined test cases. Test cases can use ontologies of different nature (from expressive OWL ontologies to simple directories) and use different modalities, eg, blind evaluation, open evaluation, consensus. OAEI-2009 builds over previous campaigns by having 5 tracks with 11 test cases followed by 16 participants. This paper is an overall presentation of the OAEI 2009 campaign.

Abstract. State of the art ontology matching techniques are limited to detect simple correspondences between atomic concepts and properties. Nevertheless, for many concepts and properties atomic counterparts will not exist, while it is possible to construct equivalent complex concept ...

With the growth of the Linked Data Web, time-efficient Link Discovery frameworks have become indispensable for implementing the fourth Linked Data principle, i.e., the provision of links between data sources. Due to the sheer size of the Data Web, detecting links even when using trivial specifications based on a single property can be very timedemanding. Moreover, non-trivial Link Discovery tasks require complex link specifications and are consequently even more challenging to optimize with respect to runtime. In this paper, we present a novel hybrid approach to link discovery that combines two very fast algorithms. Both algorithms are combined by using original insights on the translation of complex link specifications to combinations of atomic specifications via a series of operations on sets and filters. We show in three experiments that our approach outperforms SILK by more than six orders of magnitude while abiding to the restriction of not losing any link.

Current ontology matching techniques focus on detecting correspondences between atomic concepts and properties. Nevertheless, it is necessary and possible to detect correspondences between complex concept or property descriptions. In this paper, we demonstrate how complex matching can benefit from natural language processing techniques, and propose an enriched set of correspondence patterns leveraging linguistic matching conditions. After elaborating on the integration of methods for the linguistic analysis of textual labels with an existing framework for detecting complex correspondences, we present the results of an experimental evaluation on an OAEI dataset. The results of our experiments indicate a large increase of precision as compared to the original approach, which was based on similarity measures and thresholds.

State of the art ontology matching techniques are limited to detect simple correspondences between atomic concepts and properties. Nevertheless, for many concepts and properties atomic counterparts will not exist, while it is possible to construct equivalent complex concept and property descriptions. We define a correspondence where at least one of the linked entities is non-atomic as complex correspondence. Further, we introduce several patterns describing com- plex correspondences. In particular, we focus on methods for automatically de- tecting complex correspondences. These methods are based on a combination of basic matching techniques. We conduct experiments with different datasets and discuss the results.

Current ontology matching techniques focus on detecting correspondences between atomic concepts and properties. Nevertheless, it is necessary and possible to detect correspondences between complex concept or property descriptions. In this paper, we demonstrate how complex matching can benefit from natural language processing techniques, and propose an enriched set of correspondence patterns leveraging linguistic matching conditions. After elaborating on the integration of methods for the linguistic analysis of textual labels with an existing framework for detecting complex correspondences, we present the results of an experimental evaluation on an OAEI dataset. The results of our experiments indicate a large increase of precision as compared to the original approach, which was based on similarity measures and thresholds.

With the growth of the Linked Data Web, time-efficient Link Discovery frameworks have become indispensable for implementing the fourth Linked Data principle, i.e., the provision of links between data sources. Due to the sheer size of the Data Web, detecting links even when using trivial specifications based on a single property can be very timedemanding. Moreover, non-trivial Link Discovery tasks require complex link specifications and are consequently even more challenging to optimize with respect to runtime. In this paper, we present a novel hybrid approach to link discovery that combines two very fast algorithms. Both algorithms are combined by using original insights on the translation of complex link specifications to combinations of atomic specifications via a series of operations on sets and filters. We show in three experiments that our approach outperforms SILK by more than six orders of magnitude while abiding to the restriction of not losing any link.

Ontology matching 1 is a key interoperability enabler for the Semantic Web, as well as a useful tactic in some classical data integration tasks. It takes the ontologies as input and determines as output an alignment, that is, a set of correspondences between the semantically related entities of those ontologies. These correspondences can be used for various tasks, such as ontology merging and data translation. Thus, matching ontologies enables the knowledge and data expressed in the matched ontologies to interoperate.

Traditionally, research data and publications are held in separate systems. This results in a disadvantageous situation for researchers as they need to use a variety of different systems to find relevant information about a topic. We therefore face the challenge to overcome the boundaries between bibliographic records and research data by providing an integrated search environment for publications and research data. Because of the inherently different system structure and the diverse metadata for publications and datasets respectively, one type of data cannot easily be integrated into an information system designed for another data type. We present the challenges that arise when adapting a bibliographic library system to include the additional data and give recommendations for an efficient implementation. By presenting our enhanced prototype, we show the applicability and practicability of our proposed solutions. Since our library catalogue prototype features links between publications and underlying research datasets, we provide direct access to metadata of research data stored in remote research data repositories and thus connect both types of information systems.

Purpose: Sales and procurement are the main boundary spanning functions of an organization-each with a specific focus and partly different views and objectives. They are often considered as two sides of a coin that struggle with one another for relative competitive advantage. The digitalization of procurement functions and the introduction of enterprise resource systems have led to a seeming data abundance. However, the results especially in the area of artificial intelligence are not yet satisfactory in practical application. In addition, few academic works are steered towards procurement. In fact, some expect that procurement is less likely to benefit from the application of methods of artificial intelligence (AI) emphasizing the potential benefits in functions such as finance, production, marketing and sales. Why is that? What can we do about it? Or is it even a bad thing after all? Methodology: Explanatory study considering the needed decisions and available data of procurement in contrast to sales and marketing. The manuscript is structured in three sections based upon the "Memorandum of Design-Orientated Business Informatics" with analysis, draft, evaluation and diffusion (Österle et al., 2010). Findings: There is a need for research on the purchasing-marketing interface, not just for AI but also for AI applications and analytics in general. Procurement scholars and managers must speed up data and analytical development, especially since our negotiation partners are benefiting from the rapid development of AI technology. Five propositions have been derived from a master thesis analyzing the needed decision and available data as well as during discussions at the 2021 European Research Seminar to facilitate practical application for management and direct further research. These are more perceived value, more data, better technological solutions, more skills and training, and different role in the value chain. Originality: This is an original work of in-progress research. The State of Artificial Intelligence: Procurement versus Sales and Marketing 1 Analysis Artificial intelligence (AI) is a research area that attempts to design mechanisms allowing machines to develop intelligent behavior (Russell et al., 2009). In McKinsey's (Balakrishnan et al., 2020) as well as Deloitte's (Ammanath et al., 2020) "The state of AI" survey the business functions in which organizations adopt AI are over time largely consistent with service operations, product development, marketing and sales. Since sales seem to have a faster pace, procurement must speed up in the analytics race (Chae et al., 2014, Allal-Chérif et al., 2021). Why, for instance, is procurement not at large using AI for supplier risk assessment when sales is using AI for demand forecasting? For big data analytics in general, research has shown that procurement is also lagging (Nguyen et al., 2018, Vollmer et al., 2019). Searches on academic and popular databases utilizing keywords such as procurement as well as sales and marketing respectively with artificial intelligence / machine learning / big data analytics further indicate that there is more research and application in sales and marketing. In addition, there are many search hits, what salespeople can learn from buyers but interestingly not the other way around. Maybe now is the right time to turn the coin on the other side also from a procurement perspective! One example is an article in the magazine Spend Matters, which urges management to concentrate on information sharing throughout the value chain with common metrics. Based on data from the Association of National Advertisers in the United States of America in the article it is highlighted that while marketers are concerned with strengthening marketing performance, buyers are focused on reducing costs and making internal processes less risky and more efficient (Wyld, 2014). Research question: Why is procurement lagging in AI adoption versus marketing / sales? Procurement as a discipline sometimes still feels inferior to other disciplines, however, that is not per se a good enough reason to compare and postulate that procurement should be as advanced, i.e., there could be supply chain alignment reasons. Technology adoption processes, in general, have been divergent (Venkatesh et al., 2003), and no commonly accepted view exists still today. On the contrary, there are numerous models available, some of which complement each other and some that are contradictory (Mahlamäki et al., 2021). It could also be a matter of internal integration as the use of AI

Support functions such as procurement may be understood as workflow systems delivering value to the organization. A prototype simulation was built that depending on inputs such as total spend, diversity of requisitions, and company strategy-through the internal delivery organization with the people, culture, and processes-is delivering output in terms of cost savings, processing speed, and decision quality.

Procurement and marketing are the main boundary-spanning functions of an organization. Some studies highlight that procurement is less likely to benefit from artificial intelligence emphasizing its potential in other functions, i.e., in marketing. A case study in the automotive industry of the bundling problem utilizing the design science approach is conducted from the perspective of the buying organization contributing to theory and practice. We rely on information processing theory to create a practical tool that is augmenting the skills of expert buyers through a recommendation engine to make better decisions in a novel way to further save costs. Thereby, we are adding to the literature on spend analysis that has mainly been looking backward using historical data of purchasing orders and invoices to infer saving potentials in the future-our study supplements this approach with forward-looking planning data with inherent challenges of precision and information-richness.

Artificial intelligence is a research area that attempts to design mechanisms allowing machines to develop intelligent behavior. It is a key technology for procurement and its usage is still in its infancy. For instance, the Volkswagen "Procurement Strategy 2025" stresses the potential of artificial intelligence to optimize processes and structures-and this applies to the automotive industry and other procurement organizations worldwide. Yet, only a few have successfully integrated artificial intelligence methods into their operations and across their supply chains but is recently starting to emerge. This constitutes a research opportunity on how artificial intelligence increases its performance. The Ph.D. is set up as external doctoral research supported by Porsche and the Volkswagen AutoUni in cooperation with the University of Mannheim. The research goal is to examine and exploit ideas on how methods of artificial intelligence can be utilized in the procurement function. Procurement is often one of the last functions to be digitized. However, it must keep up in the race against the capabilities of our negotiation partners in the sales organizations of our suppliers worldwide.

Web mining -is the application of data mining techniques to discover patterns from the Web. Topic tracking is one of the technologies that has been developed and can be used in the text mining process. The main purpose of topic tracking is to identify and follow events presented in multiple news sources, including newswires, radio and TV broadcasts. In this paper, a survey of topic tracking techniques is presented.

Web mining is the application of data mining techniques to discover patterns from the Web. Topic tracking is one of the technologies that has been developed and can be used in the text mining process. The main purpose of topic tracking is to identify and follow events presented in multiple news sources, including newswires, radio and TV broadcasts. In this paper, a survey of topic tracking techniques is presented. KeywordsText Mining, Topic detection, topic tracking

Power has totally different attributes than other material commodities as electrical energy stockpiling is a costly phenomenon. Since it should be generated when demanded, it is necessary to forecast its demand accurately and efficiently. As electrical load data is represented through time series pattern having linear and non-linear characteristics, it needs a model that may handle this behavior well in advance. This paper presents a scalable and hybrid approach for forecasting the power load based on Vector Auto Regression (VAR) and hybrid deep learning techniques like Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN). CNN and LSTM models are well known for handling time series data. The VAR model separates the linear pattern in time series data, and CNN-LSTM is utilized to model non-linear patterns in data. CNN-LSTM works as CNN can extract complex features from electricity data, and LSTM can model temporal information in data. This approach can derive temporal ...