Yinuo Zhang

Evaluating continuous probabilistic queries over imprecise sensor data

Pervasive applications, such as natural habitat monitoring and location-based services, have attracted plenty of research interest. These applications deploy a large number of sensors (e.g. temperature sensors) and positioning devices (e.g. GPS) to collect data from external environments. Very often, these systems have limited network bandwidth and battery resources. The sensors also cannot record accurate values. The uncertainty of these data hence has to been taken into account for query evaluation purposes. In particular, probabilistic queries, which consider data impreciseness and provide statistical guarantees in answers, have been recently studied. In this paper, we investigate how to evaluate a long-standing (or continuous) probabilistic query. We propose the probabilistic filter protocol, which governs remote sensor devices to decide upon whether values collected should be reported to the query server. This protocol effectively reduces the communication and energy costs of sensor devices. We also introduce the concept of probabilistic tolerance, which allows a query user to relax answer accuracy, in order to further reduce the utilization of resources. Extensive simulations on realistic data show that our method reduces by address more than 99% of savings in communication costs.

Probabilistic filters: A stream protocol for continuous probabilistic queries

Pervasive applications, such as natural habitat monitoring and location-based services, have attracted plenty of research interest. These applications, which deploy a lot of sensor devices to collect data from external environments, often have limited network bandwidth and battery resources. The sensors also cannot record accurate values. The uncertainty of data captured by a sensor should thus be considered for query evaluation. To this end, probabilistic queries, which consider data impreciseness and provide statistical guarantees in answers, have been recently studied. We investigate the evaluation of a long-standing (or continuous) probabilistic query in a multi-user environment. We propose the probabilistic filter protocol, which helps remote sensor devices to decide whether values collected should be reported to the query server. This protocol can significantly reduce the communication and energy costs of sensor devices. We further introduce probabilistic tolerance, which allows a query user to relax answer accuracy, in order to further reduce the utilization of resources. We extend the protocol to facilitate concurrent handling of multiple user query requests. Experimental results on sensor and location data show that our method significantly reduces communication, energy consumption, and computational overhead of the system.

UFOM: Unified fuzzy ontology matching

Unified Fuzzy Ontology Matching (UFOM) is an ontology matching system designed to discover semantic links between large real-world ontologies populated with entities from heterogeneous sources. In such ontologies, several entities in different ontologies are expected to be related to each other but not necessarily with one of the typical well-defined correspondence relationships (equivalent-to, subsumed-by). In particular, we define a new kind of correspondence relation called Relevance that reflects the relation between entities when they share a certain amount of mutual information. UFOM uses fuzzy set theory as a general framework for fuzzy ontology alignment. The framework enables representation of multiple types of correspondence relations and characterization of the uncertainty in the correspondence discovery process. UFOM computes multiple measures of similarity among ontology entities - syntactic, semantic, and structural. These measures are composed in a principled manner for ontology alignment. The system is evaluated using publicly available ontologies provided by Ontology Alignment Evaluation Initiative (OAEI). The performance of the proposed system is comparable to the top performing ontology matchers in OAEI. We also evaluate the UFOM system on a dataset from an enterprise application domain.

UFOMQ: An Algorithm for Querying for Similar Individuals in Heterogeneous Ontologies

The chief challenge in identifying similar individuals across multiple ontologies is the high computational cost of evaluating similarity between every pair of entities. We present an approach to querying for similar individuals across multiple ontologies that makes use of the correspondences discovered during ontology alignment in order to reduce this cost. The query algorithm is designed using the framework of fuzzy logic and extends fuzzy ontology alignment. The algorithm identifies entities that are related to the given entity directly from a single alignment link or by following multiple alignment links. We evaluate the approach using both publicly available ontologies and from an enterprise-scale dataset. Experiments show that it is possible to trade-off a small decrease in precision of the query results with a large savings in execution time.

Enriching employee ontology for enterprises with knowledge discovery from social networks

To enhance human resource management and personalized information acquisition, employee ontology is used to model business concepts and relations between them for enterprises. In this paper, we propose an employee ontology that integrates user static properties from formal structures with dynamic interests and expertise extracted from informal communication signals. We mine users interests at both personal and professional level from informal interactions on communication platforms at the workplace. We show how complex semantic queries enable granular analysis. At the microscopic level, enterprises can utilize the results to better understand how their employees work together to complete tasks or produce innovative ideas, identify experts and influential individuals. At the macroscopic level, conclusions can be drawn, among others, about collective behavior and expertise in varying granularities (i.e. single employee to the company as a whole).

Computational Cost of Querying for Related Entities in Different Ontologies

The computational cost of querying for similar entities across ontologies is high since, in the worst case, every pair of entities will have to be considered. Therefore, links discovered during ontology alignment have been used to speed up querying across ontologies by following relatedness links to discover similar entities. We derive the computational complexity of querying across ontologies using the ontology alignment links discovered using the Unified Fuzzy Ontology Matching (UFOM) framework. We consider querying for related entities by following either a single alignment link or by following multiple alignment links. These methods have different computational complexity and produce different query results. We also study the impact of the specific implementation approach on query time. We consider implementations based on multiple accesses of the triplestore using a high-level procedural language and by execution of a single SPARQL graph query on the ontology server. These approaches were evaluated using ontologies derived from an enterprise-scale dataset. Experimental results show that an implementation using nested for-loops in a procedural language outperformed by nearly 2× an implementation based on a single SPARQL query.

Integration of Heterogeneous Web Services for Event-Based Social Networks

Event-based online social networks are Internet-based services that enable users to participate in real world experiences together. Event-based social networks can be created by a community of end-users based on their own interests in specific types of event and sources of event information. We propose a method to create such event-based social networks through integration of existing online information sources of events using a Semantic Web framework. In order to match people with common interests in such activities to self-organize into a social network, we integrate information from heterogeneous information sources related to event schedules, ticket purchases, and group attendance from multiple online sources. The Semantic Web framework is used to represent these heterogeneous datasets and unstructured online data is converted into ontologies. Links between event information in different sources are discovered using both the syntactic similarity and semantic similarity between ontology classes. We use an approach based on Latent Dirichlet Allocation (LDA) over the space of topics related to each event and user profiles for event recommendation. This enables the event-based social network to recommend friends based on shared interest in an event - online friendship is established after mutual attendance of the same event. We demonstrate this approach with EasyGo, a web-based mashup application which integrates information of events such as concerts, sports, theatres, as well as tickets and group purchase from multiple online sources.

FP-CPNNQ: A Filter-Based Protocol for Continuous Probabilistic Nearest Neighbor Query

An increasing number of applications in environmental monitoring and location-based services make use of large-scale distributed sensing provided by wireless sensor networks. In such applications, a large number of sensor devices are deployed to collect useful information such as temperature readings and vehicle positions. However, these distributed sensors usually have limited computational and communication power and thus the amount of sensor queries should be reduced to conserve system resources. At the same time, data captured by such sensors is inherently imprecise due to sensor limitations. We propose an efficient probabilistic filter-based protocol for answering continuous nearest neighbor queries over uncertain sensor data. Experimental evaluation on real-world temperature sensing data and synthetic location data showed a significant reduction in the number of update messages.

Rapid Data Integration and Analysis for Upstream Oil and Gas Applications

The increasingly large number of sensors and instruments in the oil and gas industry, along with novel means of communication in the enterprise has led to a corresponding increase in the volume of data that is recorded in various information repositories. The variety of information sources is also expanding: from traditional relational databases to time series data, social network communications, collections of unsorted text reports, and linked data available on the Web. Enabling end-to-end optimization considering these diverse types of information requires creating semantic links between them. Though integration of data across silo-ed databases has been recognized as a problem for a long time, it has proven to be difficult to accomplish due to the complexity of the data arrangement within databases, scarcity of metadata that describe the content, lack of a direct mapping between related entities across databases, and the several types of data represented within a database. In addition, there are large amounts of unstructured text data such as text entries in databases and document repositories. These contain valuable information on processes from the field but there is currently no method to convert this raw data to useable information. The Center for Interactive Smart Oilfield Technologies (CiSoft) is a USC-Chevron Center of Excellence for Research and Academic Training on Smart Oilfield Technologies. We describe the Integrated Optimization project at CiSoft which has the goal of developing a framework for automated linking of heterogeneous data sources and analysis of the integrated data in the context of upstream applications.