Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Data Mining Practical Machine Learning Tools and Techniques

How to Do Things With Words

Knowledge graphs enable a wide variety of applications, including question answering and information retrieval. Despite the great effort invested in their creation and maintenance, even the largest (e.g., Yago, DBPedia or Wikidata) remain incomplete. We introduce Relational Graph Convolutional Networks (R-GCNs) and apply them to two standard knowledge base completion tasks: Link prediction (recovery of missing facts, i.e. subject-predicate-object triples) and entity classification (recovery of missing entity attributes). R-GCNs are related to a recent class of neural networks operating on graphs, and are developed specifically to handle the highly multi-relational data characteristic of realistic knowledge bases. We demonstrate the effectiveness of R-GCNs as a stand-alone model for entity classification. We further show that factorization models for link prediction such as DistMult can be significantly improved through the use of an R-GCN encoder model to accumulate evidence over multiple inference steps in the graph, demonstrating a large improvement of 29.8% on FB15k-237 over a decoder-only baseline.

/pdf/modeling-relational-data-with-graph-convolutional-networks-81d86mzfgx.pdf

Modeling Relational Data with Graph Convolutional Networks

Ontologies tend to be found everywhere. They are viewed as the silver bullet for many applications, such as database integration, peer-to-peer systems, e-commerce, semantic web services, or social networks. However, in open or evolving systems, such as the semantic web, different parties would, in general, adopt different ontologies. Thus, merely using ontologies, like using XML, does not reduce heterogeneity: it just raises heterogeneity problems to a higher level. Euzenat and Shvaikos book is devoted to ontology matching as a solution to the semantic heterogeneity problem faced by computer systems. Ontology matching aims at finding correspondences between semantically related entities of different ontologies. These correspondences may stand for equivalence as well as other relations, such as consequence, subsumption, or disjointness, between ontology entities. Many different matching solutions have been proposed so far from various viewpoints, e.g., databases, information systems, and artificial intelligence. The second edition of Ontology Matching has been thoroughly revised and updated to reflect the most recent advances in this quickly developing area, which resulted in more than 150 pages of new content. In particular, the book includes a new chapter dedicated to the methodology for performing ontology matching. It also covers emerging topics, such as data interlinking, ontology partitioning and pruning, context-based matching, matcher tuning, alignment debugging, and user involvement in matching, to mention a few. More than 100 state-of-the-art matching systems and frameworks were reviewed. With Ontology Matching, researchers and practitioners will find a reference book that presents currently available work in a uniform framework. In particular, the work and the techniques presented in this book can be equally applied to database schema matching, catalog integration, XML schema matching and other related problems. The objectives of the book include presenting (i) the state of the art and (ii) the latest research results in ontology matching by providing a systematic and detailed account of matching techniques and matching systems from theoretical, practical and application perspectives.

Ontology Matching

Linked Open Data browsers nowadays usually list facts about entities, but they typically do not respect the relatedness of those facts. At the same time, query logs from LOD datasets hold information about which facts are typically queried in conjunction, and should thus provide a notion of intra-fact relatedness. In this paper, we examine the hypothesis how query logs can be used to improve the display of information from DBpedia, by grouping presumably related facts together. The basic assumption is that properties which frequently co-occur in SPARQL queries are highly semantically related, so that co-occurence in query logs can be used for visual grouping of statements in a Linked Data browser. A user study, however, shows that the grouped display is not significantly better than simple baselines, such as the alphabetical ordering used by the standard DBpedia Linked Data interface. A deeper analysis shows that the basic assumption can be proven wrong, i.e., co-occurrence in query logs is actually not a good proxy for semantic relatedness of statements.

/pdf/what-sparql-query-logs-tell-and-do-not-tell-about-semantic-25ajb6jr2u.pdf

What SPARQL Query Logs Tell and Do Not Tell About Semantic Relatedness in LOD

The goal of reference models is to establish a common vocabulary and recently also to facilitate semantically unambiguous information exchange between IT systems. However, IT systems are based on implementation models that typically deviate significantly from the reference models. This raises the need for a mapping mechanism, which is flexible enough to cope with the disparities between implementation model and reference model at runtime and on instance level, and which can be implemented without altering the established IT system. We present an architecture that solves this problem by establishing methods for representing the instances of an existing IT-System in terms of a reference model. Based on rules, the concrete nature of the representation is decided at run time. Albeit our approach is entirely domain independent, we demonstrate the feasibility of our approach in an industrial case study from the Oil and Gas domain, using the ISO 15926 ontology as a reference model and mapping it to different Java and Flex implementation models.

/pdf/an-architecture-for-information-exchange-based-on-reference-36lnu29al8.pdf

An architecture for information exchange based on reference models

Emergency management applications support a command staff in disruptive disaster situations, such as earthquakes, large-scale floodings or fires. One crucial requirement to emergency management systems is to provide decision makers with the relevant information to support their decisions. Mashups can help here by providing flexible and easily understandable views on up-to-date information. In this chapter, we introduce a number of mashups from the domain of emergency management. An in-depth study of the mashup MICI shows how mashups can combine valuable information for ranking and filtering emergency calls to cope with information shortage and overload. We further discuss the use of Linked Open Data both as a source of additional information and a means for more intelligent filtering.

Mashups for the Emergency Management Domain

Ontologies can be used for e-business integration, for example by describing existing e-business standards as ontologies. If cooperating parties use different ontologies, ontology mappings are needed, which can be ambiguous, thus making ontology mapping disambiguation necessary. Different disambiguation strategies exist, such as community-driven or context-sensitive referencing of ontologies, where the latter is what we developed in our project. In this paper, we show that community-driven referencing can be realized using a context-sensitive referencing service in a way that the user administration is transparent to the referencing system.

/pdf/context-sensitive-referencing-for-ontology-mapping-3piyl7vqfo.pdf

Context-Sensitive Referencing for Ontology Mapping Disambiguation

Determining the semantic relatedness (i.e., the strength of a relation) of two resources in DBpedia (or other Linked Data sources) is a problem addressed by quite a few approaches in the recent past. However, there are no large-scale benchmark datasets for comparing such approaches, and it is an open problem to determine which of the approaches work better than others. Furthermore, larget-scale datasets for training machine learning based approaches are not available. DBpedia-NYD is a large-scale synthetic silver standard benchmark dataset which supports contains symmetric and asymmetric similarity values, obtained using a web search engine.

/pdf/dbpedianyd-a-silver-standard-benchmark-dataset-for-semantic-2bs7nwzby1.pdf

Heiko Paulheim

Papers

What SPARQL Query Logs Tell and Do Not Tell About Semantic Relatedness in LOD

An architecture for information exchange based on reference models

Mashups for the Emergency Management Domain

Context-Sensitive Referencing for Ontology Mapping Disambiguation

DBpediaNYD: a silver standard benchmark dataset for semantic relatedness in DBpedia