AIM Laboratory

Introduction:

The Advanced Integration and Mining Lab (AIM Lab) was established in 2005 to conduct the data integration and data mining research at CIS of UO. There are large amount of data stored in different data repositories, such as databases, data warehouses, WWW and the emerging Semantic Web. They may use different structures (schemas) or semantics (ontologies) to describe their data even in the same domain. How to integrate those heterogenous data resources is still a big challenge to both database and Semantic Web research. Finding the relationships (mappings) is first step to do data integration and it may need human involvement. Then the mappings can be used for data translation or query answering across different data resources. Our lab is developing ontology-based information systems to help both general users and domain experts (e.g., biologists, neuroscientists) to integrate, process and analyze their data.

Data mining is a useful technique for finding interesting trends or patterns in those large datasets to guide decisions about future activities. Data mining has been used in many fields. In our lab, we have been interested in using data mining to study biomedical data, web data and network data. On the other hand, how to mine the interesting patterns (rules) across heterogenous data resources is also a very interesting topic to us. There can be some interaction between data mining and data integration tasks. For example, the generated data mining rules across different data resources can be used to guide data integration, and the integrated data can be used for finding more interesting trends and patterns.

Research Projects:

• OntoGrate is an ontology-based system for interactive information integration. The goal is to help human experts in multiple domains to interactively integrate information that is heterogenous in both structure and semantics. Key innovations in OntoGrate include broadening the typical scope of integration to span databases, XML data, Knowledge bases and the Semantic Web; strengthening and formalizing the derivation of mapping rules by introducing machine learning and data mining techniques; and extending our inference engine, OntoEngine, and first order ontology language, Web-PDDL, to solve the problem of integration using formal mapping rules. Along with OntoGrate, one sub-project is on how to use inductive logic programming to learn complex mapping rules. Another one is on how to use association rule mining to find mapping rules. The collaboration with Dr. Jongwan Kim is to construct a Personalized Ontology-based Anti-spam email System (POAS).

• Neural ElectroMagnetic Ontology (NEMO) system addresses the need for tools to support representation, storage, mining, and dissemination of brain electromagnetic (EEG and MEG) data. NEMO is a collaborative project with researchers at NIC (NeuroInformatics Center), EGI (Electrical Geodesics, Inc) and Department of Psychology from UO, and Learning Research and Development Center, University of Pittsburgh. We are developing several temporal and spatial ontologies which are novel for the EEG/MEG neuroscience field. We are are also building ontology-based temporal and spatial databases for EEG/MEG data. We expect this work to receive significant attention.

• Internet Routing Forensics (IRF) is a collaborative project with researchers in Network Security Lab and Advanced Network Technology Center. We are extending several data mining techniques, such as classification and clustering, to discover and analyze abnormal BGP (Border Gateway Protocol) events, such as worms and blackout. This project is being funded with a three year research grant (PI: Jun Li Co-PI: Dejing Dou, David Meyer) by NSF.

Software for download and online services:

We already set up a prototype of OntoGrate online service for querying relational database with Semantic Web ontologies. We will have software and complete online services be released to public finally. Currently, people can also try our inference engine, OntoEngine and OntoMerge online service for translating Semantic Web data. (Both OntoEngine and OntoMerge were firstly developed by Dejing Dou and his Yale colleague, Peishen Qi, and their advisor, Drew McDermott.)

Faculty members: