Analyzing behavioral properties of semantic business processes with parametric data

One of the basic problems in (semantic) business process management concerns the behavioral analysis at design time so that only correct processes are exposed. Although this analysis requires the semantic propagation of data and system preconditions through processes and between processes to be considered, some concrete data are usually obtained at runtime and cannot be used in early process development phases. This paper proposes Parametric Unary RDF Annotated Petri Net Systems (P‐U‐RDF‐PN Systems) as a formalism to alleviate this problem. The formalism allows analyzing behavioral properties of semantic business processes at design time. Parametric values are used to represent the data that will be obtained at runtime, allowing to also consider them in the analysis phase. The paper formally defines the semantics of the formalism in terms of enabling and firing of transitions as well as the concept of reachability graph. It also presents how to compute and analyze this parametric reachability graph. The analysis is based on an adaption of model checking techniques, being temporal logic the formalism used to specify the system properties to be analyzed. Finally, an implementation prototype using RDF and SPARQL tools and a Satisfiability Modulo Theories solver is presented, with some experiments evaluating how scalable the prototype is. Copyright © 2010 John Wiley & Sons, Ltd.     

A universal relation data model with semantic abstractions

Two important features of modern database models are support for complex data structures and support for high-level data retrieval and update. The first issue has been studied by the development of various semantic data models; the second issue has been studied through universal relation data models. How the advantages of these two approaches can be combined is presently examined. A new data model that incorporates standard concepts from semantic data models such as entities, aggregations, and ISA hierarchies is introduced. It is then shown how nonnavigational queries and updates can be interpreted in this model. The main contribution is to demonstrate how universal relation techniques can be extended to a more powerful data model. Moreover, the semantic constructs of the model allow one to eliminate many of the limitations of previous universal relation models     

Integrated machine learning with semantic web for open government data recommendation based on cloud computing

Open government data (OGD) is a type of trusted information that can be used to verify the correctness of information on social platforms. Finding interesting OGD to serve personalized needs to facilitate the development of social platforms is a challenging research topic. This study explores how to link the Taiwanese government's open data platform with Facebook and how to recommend related OGD. First, an integrated machine learning with semantic web into cloud computing framework is defined. Next, the linked data query platform (LDQP) is developed to validate its feasibility. The LDQP provides a graphical approach for personal query and links with related Facebook fan pages. LDQP automatically finds highly relevant OGD based on recent topics that users are following on Facebook when users login to Facebook via the LDQP. In this way, the LDQP query result can be dynamically adjusted and graphically displayed according to user's Facebook operations.     

A semantic object-oriented data access system

Lack of support for Entity-Relationship (E-R) semantics and the disconnect between object-oriented programming languages (OOPLs) and database languages remain key roadblocks to the effective use of object-orientation in information systems development. We present SOODAS, a Semantic Object-Oriented Data Access System that defines and manages the meta-data necessary to support E-R semantics and set level querying and provides related interface generation tools. SOODAS consists of five meta-classes. DomainObject and Relationship provide the capabilities needed to define and manage entities, attributes, relationships, external identifiers, and constraints. Together with the meta-class QueryNode, DomainObject provides an object-oriented, multi-entity querying capability. Queries can be arbitrarily complex and can include cycles and transitive closure. Persistence is provided by the meta-class, PermanentObject, of which DomainObject and Relationship are subclasses. The meta-class, DomainObjectInterface uses the meta-data in DomainObject and Relationship to generate a standard, re-usable interface for displaying and maintaining instances of any entity. Since SOODAS is implemented entirely in Smalltalk, it can be seamlessly integrated with any Smalltalk application.     

Variational multimodal machine translation with underlying semantic alignment

Capturing the underlying semantic relationships of sentences is helpful for machine translation. Variational neural machine translation approaches provide an effective way to model the uncertain underlying semantics in languages by introducing latent variables. Multitask learning is applied in multimodal machine translation to integrate multimodal data. However, these approaches usually lack a strong interpretation in utilizing out-of-text information in machine translation tasks. In this paper, we propose a novel architecture-free multimodal translation model, called variational multimodal machine translation (VMMT), under the variational framework which can model the uncertainty in languages caused by ambiguity through utilizing visual and textual information. In addition, the proposed model can eliminate the discrepancy between training and prediction in the existing variational translation models by constructing encoders only relying on source data. More importantly, the proposed multimodal translation model is designed as multitask learning in which the shared semantic representation for different modes is learned and the gap among semantic representation from various modes is reduced by incorporating additional constraints. Moreover, the information bottleneck theory is adopted in our variational encoder–decoder model, which helps the encoder to filter redundancy and the decoder to concentrate on useful information. Experiments on multimodal machine translation demonstrate that the proposed model is competitive.     

SETL: A programmable semantic extract-transform-load framework for semantic data warehouses

In order to create better decisions for business analytics, organizations increasingly use external structured, semi-structured, and unstructured data in addition to the (mostly structured) internal data. Current Extract-Transform-Load (ETL) tools are not suitable for this “open world scenario” because they do not consider semantic issues in the integration processing. Current ETL tools neither support processing semantic data nor create a semantic Data Warehouse (DW), a repository of semantically integrated data. This paper describes our programmable Semantic ETL (SETL) framework. SETL builds on Semantic Web (SW) standards and tools and supports developers by offering a number of powerful modules, classes, and methods for (dimensional and semantic) DW constructs and tasks. Thus it supports semantic data sources in addition to traditional data sources, semantic integration, and creating or publishing a semantic (multidimensional) DW in terms of a knowledge base. A comprehensive experimental evaluation comparing SETL to a solution made with traditional tools (requiring much more hand-coding) on a concrete use case, shows that SETL provides better programmer productivity, knowledge base quality, and performance.     

A semantic matching energy function for learning with multi-relational data

Large-scale relational learning becomes crucial for handling the huge amounts of structured data generated daily in many application domains ranging from computational biology or information retrieval, to natural language processing. In this paper, we present a new neural network architecture designed to embed multi-relational graphs into a flexible continuous vector space in which the original data is kept and enhanced. The network is trained to encode the semantics of these graphs in order to assign high probabilities to plausible components. We empirically show that it reaches competitive performance in link prediction on standard datasets from the literature as well as on data from a real-world knowledge base (WordNet). In addition, we present how our method can be applied to perform word-sense disambiguation in a context of open-text semantic parsing, where the goal is to learn to assign a structured meaning representation to almost any sentence of free text, demonstrating that it can scale up to tens of thousands of nodes and thousands of types of relation.     

A new multi-view learning machine with incomplete data

Pattern Analysis and Applications - Multi-view learning with incomplete views (MVL-IV) is a reliable algorithm to process incomplete datasets which consist of instances with missing views or...     

A semantic framework for textual data enrichment

In this work we present a semantic framework suitable of being used as support tool for recommender systems. Our purpose is to use the semantic information provided by a set of integrated resources to enrich texts by conducting different NLP tasks: WSD, domain classification, semantic similarities and sentiment analysis. After obtaining the textual semantic enrichment we would be able to recommend similar content or even to rate texts according to different dimensions. First of all, we describe the main characteristics of the semantic integrated resources with an exhaustive evaluation. Next, we demonstrate the usefulness of our resource in different NLP tasks and campaigns. Moreover, we present a combination of different NLP approaches that provide enough knowledge for being used as support tool for recommender systems. Finally, we illustrate a case of study with information related to movies and TV series to demonstrate that our framework works properly.     

A unified multimedia and semantic perspective for data retrieval in the semantic web

In recent years, the emerging diffusion of peer-to-peer networks is going beyond the single-domain paradigm like, for instance, the mono-thematic file sharing one (e.g. Napster for music). Peers are more and more heterogeneous data sources which need to share data with commercial, educational, and/or collaboration purposes, just to mention a few. Moreover, in current information processing applications data cannot be meaningfully searched by precise database queries that would return exact matches (e.g. when dealing with multimedia, proteomic, statistical data).     

A RESTful and semantic framework for data integration

Companies have to deal with huge amounts of heterogeneous information, usually stored in distributed datasets that make use of different data schemas. This topic is especially crucial for enterprises that deal with new and different kinds of business data as new services are provided; they need to be able to dynamically add new datasets with new schemas to their information systems. However, even though research efforts have been applied to deal with this integration problem, there is still a lack of practical approaches ready to be implemented for industrial cases. We present a web‐based architecture and system built upon ontologies and other semantic web techniques to cope with federation of business data in real time. The scenario used to demonstrate the utility of the architecture is composed of actual data of a telecom company. Results show that our solution is more suitable, efficient and practical than other works. Copyright © 2014 John Wiley & Sons, Ltd.     

A parametric model for fusing heterogeneous fuzzy data

Presented is a model that integrates three data types (numbers, intervals, and linguistic assessments). Data of these three types come from a variety of sensors. One objective of sensor-fusion models is to provide a common framework for data integration, processing, and interpretation. That is what our model does. We use a small set of artificial data to illustrate how problems as diverse as feature analysis, clustering, cluster validity, and prototype classifier design can all be formulated and attacked with standard methods once the data are converted to the generalized coordinates of our model. The effects of reparameterization on computational outputs are discussed. Numerical examples illustrate that the proposed model affords a natural way to approach problems which involve mixed data types     

基于SVM的局部潜在语义分析算法研究

针对现有的Web文本分类与表示方法中出现的各种分类效果与性能优化等问题,基于局部潜在语义分析的理论原理,利用支持向量机分类优势,设计出一种基于文档与类别之间相关度的生成局部区域的算法,即S-LLSA。该算法在语义分析使用矩阵的奇异值分解过程中引入不同类别信息,分析特征词的局部特征,使用支持向量机分类器计算文本对类别的相关度参数,并应用于局部区域生成过程。通过实验表明,S-LLSA算法有效解决了局部区域如何进行局部奇异值分解问题,有效提高并优化了Web文本分类效果,更好地表示了Web文本潜在语义空间。     

Intimate machine interaction

It is argued that current digital media offer exciting new opportunities for experimental work based on sharing responsibilities and tasks between man and machine. The artist can get a better in-deph understanding of his individual exploratory behaviour through navigating in behaviour-rich problem domains. Interactive methods are instrumental to introspection and flexible enough to control and explore many different types of dynamic systems.     

A semantic enriched data model for sensor network interoperability

One of the main open issues in the development of applications for sensor network management is the definition of interoperability mechanisms among the several monitoring systems and heterogeneous data. Interesting researches related to integration techniques have taken place, they are primary based on the adoption of sharing data mechanisms. In the last years, the Service-Oriented Architecture (SOA) approach has become predominant in many sensor network projects as it enables the cooperation and interoperability of different sensor platforms at an higher level of abstraction. In this paper we propose a novel architecture for the interoperability of sensor networks, which is based on web services technologies and on a common data model enriched with semantic concepts and annotations. The proposed architecture allows the development of complex decision support system applications by integration of heterogeneous data, accessible through services, according to standard data format and standard protocols.     

Exemplar-based statistical model for semantic parametric design of human body

This paper presents an exemplar-based method to provide intuitive way for users to generate 3D human body shape from semantic parameters. In our approach, human models and their semantic parameters are correlated as a single linear system of equations. When users input a new set of semantic parameters, a new 3D human body will be synthesized from the exemplar human bodies in the database. This approach involves simpler computation compared to non-linear methods while maintaining quality outputs. A semantic parametric design in interactive speed can be implemented easily. Furthermore, a new method is developed to quickly predict whether the parameter values is reasonable or not, with the training models in the human body database. The reconstructed human bodies in this way will all have the same topology (i.e., mesh connectivity), which facilitates the freeform design automation of human-centric products.