Open Access. Powered by Scholars. Published by Universities.®
Science and Technology Studies Commons™
Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 2 of 2
Full-Text Articles in Science and Technology Studies
Data Driven Encoding Of Structures And Link Predictions In Large Xml Document Collections, M Hagenbuchner, A C. Tsoi, M Kc, Shujia Zhang
Data Driven Encoding Of Structures And Link Predictions In Large Xml Document Collections, M Hagenbuchner, A C. Tsoi, M Kc, Shujia Zhang
Ah Chung Tsoi
In recent years there have been some significant research towards the ability of processing related data, particularly the relatedness among atomic elements in a structure with those in another structure. A number of approaches have been developed with various degrees of success. This chapter provides an overview of machine learning approaches for the encoding of related atomic elements in one structure with those in other structures. The chapter briefly reviews a number of unsupervised approaches for such data structures which can be used for solving generic classification, regression, and clustering problems. We will apply this approach to a particularly interesting …
Neural Network Classification And Prior Class Probabilities, Steve Lawrence, Ian Burns, Andrew Back, Ah Chung Tsoi, C Lee Giles
Neural Network Classification And Prior Class Probabilities, Steve Lawrence, Ian Burns, Andrew Back, Ah Chung Tsoi, C Lee Giles
Ah Chung Tsoi
A commonly encountered problem in MLP (multi-layer perceptron) classification problems is related to the prior probabilities of the individual classes - if the number of training examples that correspond to each class varies significantly between the classes, then it may be harder for the network to learn the rarer classes in some cases. Such practical experience does not match theoretical results which show that MLPs approximate Bayesian a posteriori probabilities (independent of the prior class probabilities). Our investigation of the problem shows that the difference between the theoretical and practical results lies with the assumptions made in the theory (accurate …