Fitness assessment of document model

Document classification is a supervised machine learning process, where predefined category labels are assigned to documents based on the hypothesis derived from training set of labelled documents. Documents cannot be directly interpreted by a computer system unless they have been modelled as a collection of computable features. Rogati and Yang [M. Rogati and Y. Yang, Resource selection for domain-specific cross-lingual IR, in SIGIR 2004: Proceedings of the 27th annual international conference on Research and Development in Information Retrieval, ACM Press, Sheffied: United Kingdom, pp. 154-161.] pointed out that the effectiveness of document classification system may vary in different domains. This implies that the quality of document model contributes to the effectiveness of document classification. Conventionally, model evaluation is accomplished by comparing the effectiveness scores of classifiers on model candidates. However, this kind of evaluation methods may encounter either under-fitting or over-fitting problems, because the effectiveness scores are restricted by the learning capacities of classifiers. We propose a model fitness evaluation method to determine whether a model is sufficient to distinguish positive and negative instances while still competent to provide satisfactory effectiveness with a small feature subset. Our experiments demonstrated how the fitness of models are assessed. The results of our work contribute to the researches of feature selection, dimensionality reduction and document classification.

Time weight collaborative filtering

Collaborative filtering is regarded as one of the most promising recommendation algorithms. The item-based approaches for collaborative filtering identify the similarity between two items by comparing users' ratings on them. In these approaches, ratings produced at different times are weighted equally. That is to say, changes in user purchase interest are not taken into consideration. For example, an item that was rated recently by a user should have a bigger impact on the prediction of future user behaviour than an item that was rated a long time ago. In this paper, we present a novel algorithm to compute the time weights for different items in a manner that will assign a decreasing weight to old data. More specifically, the users' purchase habits vary. Even the same user has quite different attitudes towards different items. Our proposed algorithm uses clustering to discriminate between different kinds of items. To each item cluster, we trace each user's purchase interest change and introduce a personalized decay factor according to the user own purchase behaviour. Empirical studies have shown that our new algorithm substantially improves the precision of item-based collaborative filtering without introducing higher order computational complexity.

The predicate marker li in Mauritian Creole

There is a morpheme 'li' in Mauritian Creole (MC), which is homophonous with the 3sg pronoun, and which, in the early creole, occurs frequently between the subject and the predicate in affirmative, present tense clauses. I propose that that 'li' may have originated as a resumptive pronoun, co-referential with the subject, but following the grammaticalization of new determiner elements to mark the semantic contrasts of [±definite] and singular vs. plural, 'li' has now grammaticalized into a predicate marker. Its presence is sensitive to both the nature of the predicate, and to the definiteness and specificity features of the subject NP. My analysis is within the framework of Truth Conditional Semantics, where indefinite NPs are analyzed as variables that get introduced into the discourse, and must be bound by an operator to yield a closed proposition, with a truth value. Drawing on a comparison with a cognative morpheme 'i' in Seychellois Creole, I claim that its path to grammaticalization is linked to that of the specificity marking 'la'.

α′ phase stability in Nd-Li-sialon systems

The formability and stability of the alpha-sialon (alpha') phase was investigated in multi-cation Nd-Li-sialon systems. Four samples were prepared, ranging from a pure Nd-sialon to a pure Li-sialon, with two intermediate samples being prepared with either lithium or neodymium replacing the other alpha'-stabilising additive by 20 eq.%, as to maintain an equivalent design composition in all samples. After sintering, all samples were subsequently heat treated up to 192 h at 1450 and 1300 degreesC. While significant quantities of the beta'-sialon (beta' phase were found in most samples, the high-lithium Li-Nd-sialon sample was found to be almost pure a' phase after sintering. Furthermore, the long-term stability of the a' phase on heat treatment was also found to be superior in both multi-cation samples than in either of the single-alpha'-stabilising-cation samples. This is thought to be related to improved retention of the lithium in the multi-cation systems, as much of the lithium was found to volatilise during sintering in the neodymium-free sample. (C) 2002 Elsevier Science Ltd. All rights reserved.

Yorta Yorta: The effect of changes in traditional laws and customs in native title determinations

Members of the Yorta Yorta Aboriginal Community v State of Victoria was the first case in which a claim for native title was lodged in a non-remote area of the Australian mainland which was the subject of European settlement at an early stage in Australian history - highlights the difficulties in establishing native title claims in long settled regions of Australia - a failure to recognise the strength of oral tradition in establishing Aboriginal connection with the land.