A non-trial-and-error method for lag-lead compensator design

A universal design chart for lag-lead compensators is given. This chart allows compensators to be designed without trial-and-error while meeting four specifications: steady-state error; phase margin; gain margin; and gain (or phase) crossover frequency     

Developing a Fuzzy Risk Allocation Model for PPP Projects in China

Equitable allocation of risks between the government and the private sector in concession agreement is essential to the success of public-private partnership (PPP) projects. The decision-making process, based on the established risk allocation principles expressed in linguistic terms, requires qualitative judgment and experiential knowledge of construction experts. However, it is subjective, partial, and implicit in actual application. This paper aims to develop a fuzzy synthetic evaluation model for determining an equitable risk allocation between the government and the private sector. By doing so, it assists the PPP project practitioners to transform the risk allocation principles in linguistic terms into a more usable and systematic quantitative-based analysis using fuzzy set. Twenty-three principles and influencing factors for risk allocation were identified through a comprehensive literature review. Nine critical risk allocation criteria (RACs) that evaluate the risk carrying capability of project participants were further identified, validated, and compiled based on the experts’ knowledge via face-to-face interviews. On the other hand, the weighting for each critical risk allocation criterion was determined through a two-round Delphi questionnaire survey. A set of knowledge-based fuzzy inference rules was then established to set up the membership function for the nine RACs. Based on the research findings, a fuzzy synthetic evaluation model was finally established to determine an equitable risk allocation between the government and the private sector.     

Network coding theory: An introduction

For a long time, store-and-forward had been the transport mode in network communications. In other words, information had been regarded as a commodity that only needs to be routed through the network, possibly with replication at the intermediate nodes. In the late 1990’s, a new concept called network coding fundamentally changed the way a network can be operated. Under the paradigm of network coding, information can be processed within the network for the purpose of transmission. It was demonstrated that compared with store-and-forward, the network throughput can generally be increased by employing network coding. Since then, network coding has made significant impact on different branches of information science. The impact of network coding has gone as far as mathematics, physics, and biology. This expository work aims to be an introduction to this fast-growing subject with a detailed discussion of the basic theoretical results.     

Multiplexed automated DNA sequencing directly from single bacterial colonies

Sample preparation has been one of the major bottlenecks for large-scale DNA sequencing projects in terms of time and cost. To improve sample throughput and to integrate the front-end tasks to capillary-array DNA sequencers, protocols for directly sequencing the plasmids from a single bacterial colony in fused-silica capillaries were developed. After the colony is picked, lysis is accomplished in situ in the plastic sample tube using either a thermocycler or a heating block. Upon heating, the plasmids are released while chromosomal DNA and membrane proteins are denatured and precipitate to the bottom of the tube. After adding enzyme and Sanger reagents, the resulting solution was aspirated into the reaction capillaries by a syringe pump, and cycle sequencing was initiated. No deleterious effect upon the reaction efficiency, the on-line purification system, or the capillary electrophoresis separation was observed, even though the crude lysate was used as the template. Multiplexed online DNA sequencing data from 8 parallel channels allowed base calling up to 620 bp with an accuracy of 98%. The entire system can be automatically regenerated for repeated operation. By the marriage of colony sequencing with the capillary array sequencer, both the front end and the back end of DNA sequencing are combined in a miniaturized format. This protocol will ultimately reduce the cost of sequencing to well below current levels.     

Subsystem inference representation for fuzzy systems based uponproduct-sum-gravity rule

A new representation which expresses a product-sum-gravity (PSG) inference in terms of additive and multiplicative subsystem inferences of single variable is proposed. The representation yields additional insight into the structure of a fuzzy system and produces an approximate functional characterization of its inferred output. The form of the approximating function is dictated by the choice or polynomial, sinusoidal, or other designs of subsystem inferences. With polynomial inferences, the inferred output approximates a polynomial function the order of which is dependent on the numbers of input membership functions. Explicit expressions for the function and corresponding error of approximation are readily obtained for analysis. Subsystem inferences emulating sinusoidal functions are also discussed. With proper scaling, they produce a set of orthonormal subsystem inferences. The orthonormal set points to a possible “modal” analysis of fuzzy inference and yields solution to an additive decomposable approximation problem. This work also shows that, as the numbers of input membership functions become large, a fuzzy system with PSG inference would converge toward polynomial or Fourier series expansions. The result suggests a new framework to consider fuzzy systems as universal approximators     

A multi-faceted and automatic knowledge elicitation system (MAKES) for managing unstructured information

Management of unstructured information, such as emails, is vital for supporting knowledge work in professional services. However, the conventional way for managing unstructured information is inadequate as the knowledge work and associated tasks are becoming more complex, are dynamically changing with time and involve multiple concepts. This paper attempts to address the inadequacy, deficiency and limitations of the methods presently used to elicit knowledge from masses of unstructured information. These methods rely heavily on manpower, are time consuming and costly. With the development of a multi-faceted and automatic knowledge elicitation system (MAKES) manpower, time and cost can be dramatically reduced. The MAKES integrates the processes of collecting data, classifying unstructured information, modelling knowledge flow and social network analysis, and makes all of these actions into a connected process to audit unstructured information automatically. This audit is based on specific search criteria, search keywords, and the user behaviours of the knowledge workers. The unstructured information is automatically organized, classified and presented in a multi-facet taxonomy map. New concepts and knowledge are uncovered, analyzed and updated continuously from the incoming unstructured information, using a purpose-built knowledge elicitation algorithm named self-associated concept mapping (SACM). The capability and advantages of the MAKES are demonstrated through a successful trial implementation and a verification test conducted in an electronics trading company. Encouraging results have been achieved and a number of potential advantages have been realized. The area of application in this first deployment is based on an email-intensive organization and the proposed study will contribute to the advancement of methods and tools for managing other kinds of unstructured information.     

Active hashing and its application to image and text retrieval

In recent years, hashing-based methods for large-scale similarity search have sparked considerable research interests in the data mining and machine learning communities. While unsupervised hashing-based methods have achieved promising successes for metric similarity, they cannot handle semantic similarity which is usually given in the form of labeled point pairs. To overcome this limitation, some attempts have recently been made on semi-supervised hashing which aims at learning hash functions from both metric and semantic similarity simultaneously. Existing semi-supervised hashing methods can be regarded as passive hashing since they assume that the labeled pairs are provided in advance. In this paper, we propose a novel framework, called active hashing, which can actively select the most informative labeled pairs for hash function learning. Specifically, it identifies the most informative points to label and constructs labeled pairs accordingly. Under this framework, we use data uncertainty as a measure of informativeness and develop a batch mode algorithm to speed up active selection. We empirically compare our method with a state-of-the-art passive hashing method on two benchmark data sets, showing that the proposed method can reduce labeling cost as well as overcome the limitations of passive hashing.