Lien Fu Lai

Developing a fuzzy search engine based on fuzzy ontology and semantic search

Most of existing search engines retrieve web pages by means of finding exact keywords. Traditional keyword-based search engines suffer several problems. First, synonyms and terms similar to keywords are not taken into consideration to search web pages. Users may need to input several similar keywords individually to complete a search. Second, traditional search engines treat all keywords as the same importance and cannot differentiate the importance of one keyword from that of another. Third, traditional search engines lack an applicable classification mechanism to reduce the search space and improve the search results. In this paper, we develop a fuzzy search engine, called Fuzzy-Go. First, a fuzzy ontology is constructed by using fuzzy logic to capture the similarities of terms in the ontology, which offering appropriate semantic distances between terms to accomplish the semantic search of keywords. The Fuzzy-Go search engine can thus automatically retrieve web pages that contain synonyms or terms similar to keywords. Second, users can input multiple keywords with different degrees of importance based on their needs. The totally satisfactory degree of keywords can be aggregated based on their degrees of importance and degrees of satisfaction. Third, the domain classification of web pages offers users to select the appropriate domain for searching web pages, which excludes web pages in the inappropriate domains to reduce the search space and to improve the search results.

Enhanced Parallel Loop Self-Scheduling for Heterogeneous Multi-core Cluster Systems

Recently, more and more studies investigated the is-sue of dealing with the heterogeneity problem on heterogeneous cluster systems consisting of multi-core computing nodes. Previously we have proposed a hybrid MPI and OpenMP based loop self-scheduling approach for this kind of system. The allocation functions of several well-known schemes have been modified for better performance. Though the previous approach can improve system performance significantly, in this paper we present how to enhance the speedup further. First, we exploit the thread-level parallelism on the multi-core master node. Second, we investigate how to design a loop self-scheduling scheme which is able to smartly assign a proper chunk size according to each node’s performance. At the beginning of dispatching, we prevent the slow slaves from being as-signed too many tasks. On the other hand, the master will not assign too many small chunks to slaves at the end. Experimental results show that our approach could obtain the best speedup of 1.35.

Human-Readable Rule Generator for Integrating Amino Acid Sequence Information and Stability of Mutant Proteins

Most of the bioinformatics tools developed for predicting mutant protein stability appear as a black box and the relationship between amino acid sequence/structure and stability is hidden to the users. We have addressed this problem and developed a human-readable rule generator for integrating the knowledge of amino acid sequence and experimental stability change upon single mutation. Using information about the original residue, substituted residue, and three neighboring residues, classification rules have been generated to discriminate the stabilizing and destabilizing mutants and explore the basis for experimental data. These rules are human readable, and hence, the method enhances the synergy between expert knowledge and computational system. Furthermore, the performance of the rules has been assessed on a nonredundant data set of 1,859 mutants and we obtained an accuracy of 80 percent using cross validation. The results showed that the method could be effectively used as a tool for both knowledge discovery and predicting mutant protein stability. We have developed a Web for classification rule generator and it is freely available at http://bioinformatics.myweb.hinet.net/irobot.htm.

Extending FuzzyCLIPS for parallelizing data-dependent fuzzy expert systems

FuzzyCLIPS is a rule-based programming language and it is very suitable for developing fuzzy expert systems. However, it usually requires much longer execution time than algorithmic languages such as C and Java. To address this problem, we propose a parallel version of FuzzyCLIPS to parallelize the execution of a fuzzy expert system with data dependence on a cluster system. We have designed some extended parallel syntax following the original FuzzyCLIPS style. To simplify the programming model of parallel FuzzyCLIPS, we hide, as much as possible, the tasks of parallel processing from programmers and implement them in the inference engine by using MPI, the de facto standard for parallel programming for cluster systems. Furthermore, a load balancing function has been implemented in the inference engine to adapt to the heterogeneity of computing nodes. It will intelligently allocate different amounts of workload to different computing nodes according to the results of dynamic performance monitoring. The programmer only needs to invoke the function in the program for better load balancing. To verify our design and evaluate the performance, we have implemented a human resource website. Experimental results show that the proposed parallel FuzzyCLIPS can garner a superlinear speedup and provide a more reasonable response time.

Development of knowledge-based system for predicting the stability of proteins upon point mutations

Prediction of protein stability upon amino acid substitution is an important problem in designing stable proteins. We have developed a classification rule generator for integrating the knowledge of amino acid sequence and experimental stability change upon single mutation. These rules are human readable and hence the method enhances the synergy between expert knowledge and computational system. Utilizing the information about wild type, mutant, three neighboring residues and experimentally observed stability data, we have developed a method based on decision tree for discriminating the stabilizing and destabilizing mutants and predicting the protein stability changes upon single point mutations, which showed an accuracy of 82% and a correlation of 0.70, respectively. In addition, we have developed a fuzzy query method to predict protein stability with partial information. We have developed a web server for predicting the protein stability changes upon single mutations by using fuzzy query mechanism and it is available at http://bioinformatics.myweb.hinet.net/fqstab.htm.

A Fuzzy Query Mechanism for Human Resource Websites

Users preferences often contain imprecision and uncertainty that are difficult for traditional human resource websites to deal with. In this paper, we apply the fuzzy logic theory to develop a fuzzy query mechanism for human resource websites. First, a storing mechanism is proposed to store fuzzy data into conventional database management systems without modifying DBMS models. Second, a fuzzy query language is proposed for users to make fuzzy queries on fuzzy databases. Users fuzzy requirement can be expressed by a fuzzy query which consists of a set of fuzzy conditions. Third, each fuzzy condition associates with a fuzzy importance to differentiate between fuzzy conditions according to their degrees of importance. Fourth, the fuzzy weighted average is utilized to aggregate all fuzzy conditions based on their degrees of importance and degrees of matching. Through the mutual compensation of all fuzzy conditions, the ordering of query results can be obtained according to users preference.

Sequence Based Prediction of Protein Mutant Stability and Discrimination of Thermophilic Proteins

Prediction of protein stability upon amino acid substitution and discrimination of thermophilic proteins from mesophilic ones are important problems in designing stable proteins. We have developed a classification rule generator using the information about wild-type, mutant, three neighboring residues and experimentally observed stability data. Utilizing the rules, we have developed a method based on decision tree for discriminating the stabilizing and destabilizing mutants and predicting protein stability changes upon single point mutations, which showed an accuracy of 82% and a correlation of 0.70, respectively. In addition, we have systematically analyzed the characteristic features of amino acid residues in 3075 mesophilic and 1609 thermophilic proteins belonging to 9 and 15 families, respectively, and developed methods for discriminating them. The method based on neural network could discrimi-nate them at the 5-fold cross-validation accuracy of 89% in a dataset of 4684 proteins and 91% in a test set of 707 proteins.

High throughput computing to improve efficiency of predicting protein stability change upon mutation

Predicting protein stability change upon mutation is important for protein design. Although several methods have been proposed to improve prediction accuracy it will be difficult to employ those methods when the required input information is incomplete. In this work, we integrated a fuzzy query model based on the knowledge-based approach to overcome this problem, and then we proposed a high throughput computing method based on parallel technologies in emerging cluster or grid systems to discriminate stability change. To improve the load balance of heterogeneous computing power in cluster and grid nodes, a variety of self-scheduling schemes have been implemented. Further, we have tested the method by performing different analyses and the results showed that the present method can process hundreds of predication queries in more reasonable response time and perform a super linear speedup to a maximum of 86.2 times. We have also established a website tool to implement the proposed method and it is available at http://bioinformatics.myweb.hinet.net/para.htm.

First report of knowledge discovery in predicting protein folding rate change upon single mutation

To explore the mechanism of protein folding is one of the important topics in protein research. The accurate prediction of protein folding rate change is helpful and useful in protein design. In earlier study, we have firstly analyzed the prediction of folding rate change upon single point mutation and constructed a non-redundant dataset of F467. F467 consists of 467 mutants with various features and widely distributed on secondary structure, solvent accessibility, conservation score and long-range contacts. In this work, we therefore focused on effectively developing the knowledge in F467 dataset. We have systematically analyzed the dataset and presented several representative data mining techniques, including decision tree, decision table and association rule algorithms. Furthermore, we have interpreted, evaluated, and compared the knowledge obtained from different techniques. The experimental results showed that the present approach can effectively develop the knowledge in the dataset and the outcomes can increase the understanding of predicting protein folding rate change upon single mutation. We have also created a website with related information about this work and it is freely available at http://bioinformatics.myweb.hinet.net/kdfreedom.htm .

A Fuzzy Query Approach to Human Resource Web Services

Most human resource websites apply SQL queries to find jobs or candidates, and users must state definite conditions to conduct database queries. We apply fuzzy logic theory and the service-oriented architecture to develop human resource web services. We propose a storing mechanism store fuzzy data into conventional database management systems without modifying DBMS models and a fuzzy query language for users to make fuzzy queries on fuzzy databases. Each fuzzy condition is related to a fuzzy importance to differentiate between fuzzy conditions according to their degrees of importance. We use the fuzzy weighted average to aggregate all fuzzy conditions based on their degrees of importance and degrees of matching. Finally, our proposed service-oriented architecture simultaneously makes fuzzy queries on several human resource websites to integrate the query results.