An intelligent web-page classifier with fair feature-subset selection

The explosion of on-line information has given rise to many manually constructed topic hierarchies (such as Yahoo!!). But with the current growth rate in the amount of information, manual classification in topic hierarchies results in an immense information bottleneck. Therefore, developing an automatic classifier is an urgent need. However, classifiers suffer from enormous dimensionality, since the dimensionality is determined by the number of distinct keywords in a document corpus. More seriously, most classifiers are either working slowly or they are constructed subjectively without any learning ability. In this paper, we address these problems with a fair feature-subset selection (FFSS) algorithm and an adaptive fuzzy learning network (AFLN) for classification. The FFSS algorithm is used to reduce the enormous dimensionality. It not only gives fair treatment to each category but also has ability to identify useful features, including both positive and negative features. On the other hand, the AFLN provides extremely fast learning ability to model the uncertain behavior for classification so as to correct the fuzzy matrix automatically. Experimental results show that both FFSS algorithm and the AFLN lead to a significant improvement in document classification, compared to alternative approaches.     

大把捆远程维护仪的研究

采用单片机开发大把捆远程维护仪，通过电话线，维护仪远程拨叫维护中心，远程采集机器运行状态，远程调整参数，使机器处于最佳状态运行；维修中心专家远程判断故障点，远程维护或远程指导维护，节省维护成本和人力资源。     

Simultaneous translation of lectures and speeches

With increasing globalization, communication across language and cultural boundaries is becoming an essential requirement of doing business, delivering education, and providing public services. Due to the considerable cost of human translation services, only a small fraction of text documents and an even smaller percentage of spoken encounters, such as international meetings and conferences, are translated, with most resorting to the use of a common language (e.g. English) or not taking place at all. Technology may provide a potentially revolutionary way out if real-time, domain-independent, simultaneous speech translation can be realized. In this paper, we present a simultaneous speech translation system based on statistical recognition and translation technology. We discuss the technology, various system improvements and propose mechanisms for user-friendly delivery of the result. Over extensive component and end-to-end system evaluations and comparisons with human translation performance, we conclude that machines can already deliver comprehensible simultaneous translation output. Moreover, while machine performance is affected by recognition errors (and thus can be improved), human performance is limited by the cognitive challenge of performing the task in real time.     

Three perspectives of data mining

This paper reviews three recent books on data mining written from three different perspectives, i.e., databases, machine learning, and statistics. Although the exploration in this paper is suggestive instead of conclusive, it reveals that besides some common properties, different perspectives lay strong emphases on different aspects of data mining. The emphasis of the database perspective is on efficiency because this perspective strongly concerns the whole discovery process and huge data volume. The emphasis of the machine learning perspective is on effectiveness because this perspective is heavily attracted by substantive heuristics working well in data analysis although they may not always be useful. As for the statistics perspective, its emphasis is on validity because this perspective cares much for mathematical soundness behind mining methods.     

PAC-learnability of probabilistic deterministic finite state automata in terms of variation distance

We consider the problem of PAC-learning distributions over strings, represented by probabilistic deterministic finite automata (PDFAs). PDFAs are a probabilistic model for the generation of strings of symbols, that have been used in the context of speech and handwriting recognition, and bioinformatics. Recent work on learning PDFAs from random examples has used the KL-divergence as the error measure; here we use the variation distance. We build on recent work by Clark and Thollard, and show that the use of the variation distance allows simplifications to be made to the algorithms, and also a strengthening of the results; in particular that using the variation distance, we obtain polynomial sample size bounds that are independent of the expected length of strings.     

Knowledge-based diagnosis of drill conditions

One major bottleneck in the automation of the drilling process by robots in the aerospace industry is drill condition monitoring. This paper describes a system approach to solve this problem through the advancement of new machine design, sensor instrumentation, metal-cutting research, and intelligent software development. All drill failures can be detected and distinguished: chisel edge wear, flank wear, crater wear, margin wear, corner wear, breakage, asymmetry, lip height difference, and chipping at lips. However, in the real manufacturing environment, different workpiece materials, drill size, drill geometry, drill material, cutting speed, feed rate, etc. will change the criteria for judging the drill condition. The knowledge base used for diagnosing the drill failures requires a huge data bank and prior exhaustive testing. A self-learning scheme is therefore introduced to the machine in order to acquire the threshold history needed for automatic diagnosis by using the same new tool under the same drilling conditions.     

试论E62型精梳机

提出E7／5模式，详述了E62型精梳机在E7／5型精梳机的基础上的改进、利用计算机辅助工程优化设计进行优化组合的成功之处以及管理、维护方面的建议，指出了E62型精梳机的九项缺陷，建议有关纺机厂避免同一模式重复性竞争，开发有自主知识产权的精梳机。     

Application of the blind source separation method to feature extraction of machine sound signals

As the result of vibration emission in air, a machine sound signal carries important information about the working condition of machinery. But in practice, the sound signal is typically received with a very low signal-to-noise ratio. To obtain features of the original sound signal, uncorrelated sound signals must be removed and the wavelet coefficients related to fault condition must be retrieved. In this paper, the blind source separation technique is used to recover the wavelet coefficients of a monitored source from complex observed signals. Since in the proposed blind source separation (BSS) algorithms it is generally assumed that the number of sources is known, the Gerschgorin disk estimator method is introduced to determine the number of sound sources before applying the BSS method. This method can estimate the number of sound sources under non-Gaussian and non-white noise conditions. Then, the partial singular value analysis method is used to select these significant observations for BSS analysis. This method ensures that signals are separated with the smallest distortion. Afterwards, the time-frequency separation algorithm, converted to a suitable BSS algorithm for the separation of a non-stationary signal, is introduced. The transfer channel between observations and sources and the wavelet coefficients of the source signals can be blindly identified via this algorithm. The reconstructed wavelet coefficients can be used for diagnosis. Finally, the separation results obtained from the observed signals recorded in a semi-anechoic chamber demonstrate the effectiveness of the presented methods .     

跨世纪的测井找油找气高技术

人类社会即将跨入21世纪,各行各业都在预测未来技术的发展,测井行业也不例外。预测成象测井技术将是跨世纪的测井找油找气高技术。     

A parametric interpolator with confined chord errors, acceleration and deceleration for NC machining

Parametric interpolation has many advantages over linear interpolation in machining curves. Real time parametric interpolation research so far has addressed achieving a uniform feed rate, confined chord errors and jerk limited trajectory planning. However, simultaneous consideration of confined chord errors that respect the acceleration and deceleration capabilities of the machine has not been attempted. In this paper, the offline detection of feed rate sensitive corners is proposed. The velocity profile in these zones is planned so that chord errors are satisfied while simultaneously accommodating the machine's acceleration and deceleration limits. Outside the zone of the feed rate sensitive corners, the feed rate is planned using the Taylor approximation. Simulation results indicate that the offline detection of feed rate sensitive corners improves parametric interpolation. For real time interpolation, the parametric curve information can be augmented with the detected feed rate sensitive corners that are stored in 2×2 matrices.