Exploring Effective Methods for On-line Societal Risk Classification and Feature Mining

To identify the societal risk category of the posts of Tianya Club, several studies are carried out toward the posts of Tianya Club. With 2-month manually risk labeled new posts published during December of 2011 to January of 2012, statistical analysis of posts is conducted at first. Later, similarity analysis of posts from one risk category, different risk categories and published on different days are implemented. Finally, multi-class classification of posts using support vector machine (SVM) with different training set is tested. The statistical analysis and similarity analysis reveals the difficulties in multi-class classification of the posts of Tianya Club. The multi-class predictive results indicate that SVM could be applied to multi-class classification of posts, but still need further exploitation.

Download Full-text

The on-line use of histogram data for high-speed correction and alignment of electron microscope images

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100112713 ◽

1984 ◽

Vol 42 ◽

pp. 556-557

Author(s):

William Krakow

Keyword(s):

Power Spectrum ◽

High Speed ◽

Direct Memory Access ◽

Digital Television ◽

Contrast Transfer Function ◽

The Past ◽

High Resolution Tem ◽

On Line ◽

Correction Of Astigmatism ◽

The Fourier Transform

In the past few years on-line digital television frame store devices coupled to computers have been employed to attempt to measure the microscope parameters of defocus and astigmatism. The ultimate goal of such tasks is to fully adjust the operating parameters of the microscope and obtain an optimum image for viewing in terms of its information content. The initial approach to this problem, for high resolution TEM imaging, was to obtain the power spectrum from the Fourier transform of an image, find the contrast transfer function oscillation maxima, and subsequently correct the image. This technique requires a fast computer, a direct memory access device and even an array processor to accomplish these tasks on limited size arrays in a few seconds per image. It is not clear that the power spectrum could be used for more than defocus correction since the correction of astigmatism is a formidable problem of pattern recognition.

Download Full-text

Image registration with a computer driven S.T.E.M.

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100107630 ◽

1978 ◽

Vol 36 (1) ◽

pp. 98-99

Author(s):

A.M.H. Schepman ◽

J.A.P. van der Voort ◽

J.E. Mellema

Keyword(s):

Spot Size ◽

Scanning Transmission Electron Microscope ◽

Small Computer ◽

Structural Studies ◽

Area Of Interest ◽

Transmission Electron ◽

Scanning Transmission ◽

Specimen Area ◽

On Line ◽

Minimum Distances

A Scanning Transmission Electron Microscope (STEM) was coupled to a small computer. The system (see Fig. 1) has been built using a Philips EM400, equipped with a scanning attachment and a DEC PDP11/34 computer with 34K memory. The gun (Fig. 2) consists of a continuously renewed tip of radius 0.2 to 0.4 μm of a tungsten wire heated just below its melting point by a focussed laser beam (1). On-line operation procedures were developped aiming at the reduction of the amount of radiation of the specimen area of interest, while selecting the various imaging parameters and upon registration of the information content. Whereas the theoretical limiting spot size is 0.75 nm (2), routine resolution checks showed minimum distances in the order 1.2 to 1.5 nm between corresponding intensity maxima in successive scans. This value is sufficient for structural studies of regular biological material to test the performance of STEM over high resolution CTEM.

Download Full-text