Tyng-Ruey Chuang

Generic validation of structural content with parametric modules

We demonstrate a natural mapping from XML element types to ML module expressions. The mapping is inductive and definitions of common XML operations can be derived as the module expressions are composed. We show how to derive, in a generic way, the validation function, which checks an XML document for conformance to its DTD (Document Type Definition). One can view validation as assigning ML types to XML elements and the validation procedure a pre-requisite for typeful XML programming in ML. Our mapping uses the parametric module facility of ML in some contrived way. For example, in validating WML (WAP Markup Language) documents, we need to use 36ary type constructors, as well as higher-order modules that take in as many as 17 modules as input. That one can systematically model XML DTDs at the module level suggests ML-like languages are suitable for type-safe prototyping of DTD-aware XML applications.

Parallel Sparse Supports for Array Intrinsic Functions of Fortran 90

Fortran 90 provides a rich set of array intrinsic functions. Each of these array intrinsic functions operates on the elements of multi-dimensional array objects concurrently. They provide a rich source of parallelism and play an increasingly important role in automatic support of data parallel programming. However, there is no such support if these intrinsic functions are applied to sparse data sets. In this paper, we address this open gap by presenting an efficient library for parallel sparse computations with Fortran 90 array intrinsic operations. Our method provides both compression schemes and distribution schemes on distributed memory environments applicable to higher-dimensional sparse arrays. This way, programmers need not worry about low-level system details when developing sparse applications. Sparse programs can be expressed concisely using array expressions, and parallelized with the help of our library. Our sparse libraries are built for array intrinsics of Fortran 90, and they include an extensive set of array operations such as CSHIFT, EOSHIFT, MATMUL, MERGE, PACK, SUM, RESHAPE, SPREAD, TRANSPOSE, UNPACK, and section moves. Our work, to our best knowledge, is the first work to give sparse and parallel sparse supports for array intrinsics of Fortran 90. In addition, we provide a complete complexity analysis for our sparse implementation. The complexity of our algorithms is in proportion to the number of nonzero elements in the arrays, and that is consistent with the conventional design criteria for sparse algorithms and data structures. Our current testbed is an IBM SP2 workstation cluster. Preliminary experimental results with numerical routines, numerical applications, and data-intensive applications related to OLAP (on-line analytical processing) show that our approach is promising in speeding up sparse matrix computations on both sequential and distributed memory environments if the programs are expressed with Fortran 90 array expressions.

Real-time deques, multihead Turing machines, and purely functional programming

We answer the following question: Can a deque (doubleended queue) be implemented in a purely functional language such that each push or pop operation on either end of a queue is accomplished in 0(1) time in the worst case? The answer is yes, thus solving a problem posted by Gajewska and Tarjan [14] and by Ponder, McGeer, and Ng [25], and refining results of Sarnak [26] and Hoogerwoord [18]. We term such a deque real-time, since its constant worstcaae behavior might be useful in real time programs (assuming real-time garbage collection [3], etc.) Furthermore, we show that no restriction of the functional language is necessary, and that push and pop operations on previou

Social tagging, online communication, and Peircean semiotics: a conceptual framework

versions of a deque can also be achieved in constant time. We present a purely functional implementation of realtime deques and its complexity analysis. We then show that the implementation has some interesting implications, and can be used to give a real-time simulation of a multihead Turing machine in a purely functional language.

Building GML-native web-based geographic information systems

One of the recent web developments has focused on the opportunities it presents for social tagging through user participation and collaboration. As a result, social tagging has changed the traditional online communication process. The interpretation of tagging between humans and machines may create new problems if essential questions about how social tagging corresponds to online communications, what objects the tags refer to, who the interpreters are, and why they are engaged are not explored systematically. Since such reasoning is an interpretation of social tagging among humans, tags and machines, it is a complex issue that calls for deep reflection. In this paper, we investigate the relevance of the potential problems raised by social tagging through the framework of C.S. Peirces semiotics. We find that general phenomena of social tagging can be well classified by Peirces 10 classes of signs for reasoning. This suggests that regarding social tagging as a sign and systematically analyzing the interpretation are positively associated with the 10 classes of signs. Peircean semiotics can be used to examine the dynamics and determinants of tagging; hence, the various uses of this categorization schema may have implications for the design and development of information systems and web applications.

Browsing newsgroups with a social network analyzer

Disaster response systems are designed to facilitate decision-making based on large amounts of heterogeneous geographic information. Most geographic information systems (GISs) use relational databases to manipulate information efficiently. However, they suffer from interoperability issues because they need to expend significant effort mapping heterogeneous geographic information, which may have complicated structures, into relational data models, and vice versa. Geography Markup Language (GML) is regarded as a standard for expressing, storing, and exchanging geospatial data, and has been applied to help solve interoperability problems. Interestingly, no GIS has been built on native XML/GML technologies so far. There are two possible reasons for this: current XML processors are incapable of processing geospatial information, and they are inefficient in manipulating large XML documents. In this paper, we resolve these two difficulties and move forward to realizing GML-native web-based geographic information systems.

Non-intrusive object introspection in C++: architecture and application

Computer-mediated communication (CMC) systems such as Usenet newsgroups, e-mail lists, online chats, etc. facilitate large-scale human interactions and produce large sets of activity records. These datasets can be used to analyze and visualize the social networks existing at the online forums. Such analyses often improve our understanding of online inter-personal interactions, as well as lead to better designs and implementations of CMC systems. We report in this paper our experience in designing a newsgroup browser based on ideas from social network analysis. Our analyzer identifies reciprocal sub-groups in newsgroup discussions, and utilizes visual methods to help users look for leading authors and provocative articles. This is especially helpful for latecomers to catch up and participate in popular and noisy newsgroups.

Parallel divide‐and‐conquer scheme for 2D Delaunay triangulation

We describe the design and implementation of system architecture to support object introspection in C++. In this system, information is collected by parsing class declarations, and used to build a supporting environment for object introspection. Our approach is nonintrusive because it requires no change to the original class declarations and libraries, and it guarantees compatibility between objects before and after the addition of introspective capability. This is critical if one wants to integrate third-party class libraries, which are often supplied as black boxes and allow no modification, into highly dynamic applications. We show two applications: automatic I/O support for C++ objects; and an interactive exercise of dynamically loaded C++ class libraries.

JGAP: a Java-based graph algorithms platform

This work describes a parallel divide‐and‐conquer Delaunay triangulation scheme. This algorithm finds the affected zone, which covers the triangulation and may be modified when two sub‐block triangulations are merged. Finding the affected zone can reduce the amount of data required to be transmitted between processors. The time complexity of the divide‐and‐conquer scheme remains O(n log n), and the affected region can be located in O(n) time steps, where n denotes the number of points. The code was implemented with C, FORTRAN and MPI, making it portable to many computer systems. Experimental results on an IBM SP2 show that a parallel efficiency of 44–95% for general distributions can be attained on a 16‐node distributed memory system. Copyright

Prefiltering techniques for efficient XML document processing

We describe JGAP, a web‐based platform for designing and implementing Java‐coded graph algorithms. The platform contains a library of common data structures for implementing graph algorithms, features a ‘plug‐and‐play’ modular design for adding new algorithm modules, and includes a performance meter to measure the execution time of implemented algorithms. JGAP is also equipped with a graph editor to generate and modify graphs to have specific properties. JGAPs graphic user interface further allows users to compose, in a functional way, computation sequences from existing algorithm modules so that output from an algorithm is used as input for another algorithm. Hence, JGAP can be viewed as a visual graph calculator for helping experiment with and teach graph algorithm design. Copyright