Bayesian and Decision Tree Approaches for Pattern Recognition Including Feature Measurement Costs

The minimum cost classifier when general cost functionsare associated with the tasks of feature measurement and classification is formulated as a decision graph which does not reject class labels at intermediate stages. Noting its complexities, a heuristic procedure to simplify this scheme to a binary decision tree is presented. The optimizationof the binary tree in this context is carried out using ynamicprogramming. This technique is applied to the voiced-unvoiced-silence classification in speech processing.

Fast Public Key Cryptosystem Based On Matrix Rings

A public key cryptosystem is proposed, which is based on the assumption that finding the square root of an element in a large finite ring is computationally infeasible in the absence of a knowledge of the ring structure. The encryption and decryption operations are very fast, and the data expansion is 1:2.

Tree species diversity and ecosystem function: Can tropical multi-species plantations generate greater productivity?

Results from the humid tropics of Australia demonstrate that diverse plantations can achieve greater productivity than monocultures. We found that increases in both the observed species number and the effective species richness were significantly related to increased levels of productivity as measured by stand basal area or mean individual tree basal area. Four of five plantation species were more productive in mixtures with other species than in monocultures, offering on average, a 55% increase in mean tree basal area. A general linear model suggests that species richness had a significant effect on mean individual tree basal area when environmental variables were included in the model. As monoculture plantations are currently the preferred reforestation method throughout the tropics these results suggest that significant productivity and ecological gains could be made if multi-species plantations are more broadly pursued.

Achievements in forest tree improvement in Australia and New Zealand 6: Genetic improvement and conservation of Araucaria cunninghamii in Queensland

Araucaria cunninghamii (hoop pine) typically occurs as an emergent tree over subtropical and tropical rainforests, in a discontinuous distribution that extends from West Irian Jaya at about 0°30'S, through the highlands of Indonesian New Guinea and Papua New Guinea, along the east coast of Australia from 11°39'S in Queensland to 30°35'S in northern New South Wales. Plantations established in Queensland since the 1920s now total about 44000 ha, and constitute the primary source for the continuing supply of hoop pine quality timber and pulpwood, with a sustainable harvest exceeding 440 000 m3 y-1. Establishment of these managed plantations allowed logging of all native forests of Araucaria species (hoop pine and bunya pine, A. bidwillii) on state-owned lands to cease in the late 1980s, and the preservation of large areas of araucarian forest types within a system of state-owned and managed reserves. The successful plantation program with this species has been strongly supported by genetic improvement activities since the late 1940s - through knowledge of provenance variation and reproductive biology, the provision of reliable sources of improved seed, and the capture of substantial genetic gains in traits of economic importance (for example growth, stem straightness, internode length and spiral grain). As such, hoop pine is one of the few tropical tree species that, for more than half a century, has been the subject of continuous genetic improvement. The history of commercialisation and genetic improvement of hoop pine provides an excellent example of the dual economic and conservation benefits that may be obtained in tropical tree species through the integration of gene conservation and genetic improvement with commercial plantation development. This paper outlines the natural distribution and reproductive biology of hoop pine, describes the major achievements of the genetic improvement program in Queensland over the past 50+ y, summarises current understanding of the genetic variation and control of key selection traits, and outlines the means by which genetic diversity in the species is being conserved.

Characterization of tree-to-tree variation in morphological, nutritional and medicinal properties of Canarium indicum nuts

As part of a feasibility study of the commercialization potential of C. indicum nuts as Agroforestry Tree Products in Papua New Guinea, preliminary characterization studies have examined the tree-to-tree variation in morphological traits (nut and kernel mass and kernel:nut ratio), as well as nutritional (carbohydrate, fat, protein, sodium, vitamin E) and medicinal traits (anti-oxidant activity, anti-inflammatory activity and phenolic content) of kernels from 18 to 72 trees in a small number of different villages of Papua New Guinea (East New Britain Province). There was continuous variation in these traits indicating opportunities for multiple trait cultivar development targeted at food and pharmaceutical markets. Certain traits, for example anti-inflammatory activity, in which tree-to-tree variation was highly significant, present greater opportunities than others, such as saturated:unsaturated fatty acid ratio. This intraspecific variation was greater within populations than between populations. The data presented has allowed the development of a strategy to domesticate C. indicum for cultivation in homegardens and cocoa-coconut agroforests, using a participatory approach aimed at the production of agroforestry tree products (AFTPs) to empower small-holders and enhance their livelihoods and income.

Light environment and tree development of young Acacia melanoxylon in mixed-species regrowth forest, Tasmania, Australia

Blackwood (Acacia melanoxylon R. Br.) is a valuable leguminous cabinetwood species which is commonly found as a canopy or subcanopy tree in a broad range of mixed-species moist forests on tablelands and coastal escarpments in eastern Australia. This paper reports on the competitive light environment of a commercially valuable multi-species regrowth forest in NW Tasmania, in order to define some of the functional interactions and competitive dynamics of these stands. Comparative observations were made of the internal forest light environment in response to small-gap silvicultural treatments, in a young regenerative mix of three codominant tree species. Light measurements were made during periods of maximum external irradiance of the regrowth Eucalyptus obliqua/A. melanoxylon forest canopy at age 10.5 years. This was at a time of vigourous stand development, 4.5 years following the application of three experimental silvicultural treatments whose effects were observed in comparison with an untreated canopy sample designed as a control. Minimal irradiance was observed within and beneath the dense subcanopy of the native nurse species (Pomaderris apetala) which closely surrounds young blackwood regeneration. Unlike current plantation nurse systems, the dense foliage of the native broadleaved Pomaderris all but eliminated direct side-light and low-angle illumination of the young blackwood, from the beginning of tree establishment. The results demonstrated that retention of these densely stocked native codominants effectively suppressed both size and frequency of blackwood branches on the lower bole, through effective and persistent interception of sunlight. Vigorous young blackwood crowns later overtopped the codominant nurse species, achieving a predictable height of branch-free bole. This competitive outcome offers a valuable tool for management of blackwood crown dynamics, stem form and branch habit through manipulation of light environment in young native regrowth systems. Results demonstrate that effective self-pruning in the lower bole of blackwood is achieved through a marked reduction in direct and diffuse sunlight incident on the lower crown, notably to less than 10-15% of full sunlight intensity during conditions of maximum insolation. The results also contain insights for the improved design of mixed-species plantation nurse systems using these or functionally similar species' combinations. Based on evidence presented here for native regrowth forest, plantation nurse systems for blackwood will need to achieve 85-90% interception of external side-light during early years of tree development if self-pruning is to emulate the results achieved in the native nurse system.

Between-tree variation in fruit quality and fruit mineral concentrations of Hass avocados

Inconsistent internal fruit quality in Hass avocados affects consumer confidence. To determine the influence of individual trees on fruit quality, Hass avocado fruit were harvested from adjacent trees of similar external appearance in 3 commercial orchards in 1998 and 1 orchard in 1999. The trees in each orchard were grown with similar commercial practices and in similar soil types. Within each location, there were significant (P<0.05) differences in the mean ripe fruit quality between trees with respect to fruit body rot severity (mainly anthracnose) with and without cold storage, internal disorders severity due to diffuse discolouration and vascular browning (after cold storage), days to ripen, percentage dry matter, and the percentage of the skin area with purple-black colour when ripe. These effects were also noted in the same orchard in 1999. There were significant (P<0.05) differences in fruit flesh calcium, magnesium, potassium, boron and zinc concentrations between trees. Significant (P<0.05) correlations were observed between average fruit mineral concentrations in each tree (particularly calcium, magnesium and potassium) and body rot severity, percentage dry matter and fruit mass. There was little conclusive evidence that characteristics such as the growth of the non-suberised roots or the degree of scion under- or overgrowth was involved in these tree effects; however, differences between trees with respect to other rootstock characteristics may be involved. The inconsistency of the correlations across sites and years suggested that other factors apart from tree influences could also affect the relationship between fruit minerals and fruit quality.

Growth and physiological response of six Australian rainforest tree species to a light gradient

An understanding of growth and photosynthetic potential of subtropical rainforest species to variations in light environment can be useful for determining the sequence of species introductions in rainforest restoration projects and mixed species plantations. We examined the growth and physiology of six Australian subtropical rainforest tree species in a greenhouse consisting of three artificial light environments (10%, 30%, and 60% full sunlight). Morphological responses followed the typical sun-shade dichotomy, with early and late secondary species (Elaeocarpus grandis, Flindersia brayleyana, Flindersia schottiana, and Gmelina leichhardtii) displaying higher relative growth rate (RGR) compared to mature stage species (Cryptocarya erythroxyion and Heritiera trifoliolatum). Growth and photosynthetic performance of most species reached a maximum in 30-60% full sunlight. Physiological responses provided limited evidence of a distinct dichotomy between early and late successional species. E. grandis and F brayleyana, provided a clear representation of early successional species, with marked increase in Am in high light and an ability to down regulate photosynthetic machinery in low light conditions. The remaining species (F. schottiana, G. leichhardtii, and H. trifoliolatum) were better represented as failing along a shade-tolerant continuum, with limited ability to adjust physiologically to an increase or decrease in light, maintaining similar A(max) across all light environments. Results show that most species belong to a shade-tolerant constituency, with an ability to grow and persist across a wide range of light environments. The species offer a wide range of potential planting scenarios and silvicultural options, with ample potential to achieve rapid canopy closure and rainforest restoration goals.

Modelling predicts positive and negative interactions between three Australian tropical tree species in monoculture and binary mixture

Computer modelling promises to be an important tool for analysing and predicting interactions between trees within mixed species forest plantations. This study explored the use of an individual-based mechanistic model as a predictive tool for designing mixed species plantations of Australian tropical trees. The `spatially explicit individually based-forest simulator' (SeXI-FS) modelling system was used to describe the spatial interaction of individual tree crowns within a binary mixed-species experiment. The three-dimensional model was developed and verified with field data from three forest tree species grown in tropical Australia. The model predicted the interactions within monocultures and binary mixtures of Flindersia brayleyana, Eucalyptus pellita and Elaeocarpus grandis, accounting for an average of 42% of the growth variation exhibited by species in different treatments. The model requires only structural dimensions and shade tolerance as species parameters. By modelling interactions in existing tree mixtures, the model predicted both increases and reductions in the growth of mixtures (up to +/-50% of stem volume at 7 years) compared to monocultures. This modelling approach may be useful for designing mixed tree plantations.

Annual forecasting of the Australian macadamia crop - integrating tree census data with statistical climate-adjustment models

To facilitate marketing and export, the Australian macadamia industry requires accurate crop forecasts. Each year, two levels of crop predictions are produced for this industry. The first is an overall longer-term forecast based on tree census data of growers in the Australian Macadamia Society (AMS). This data set currently accounts for around 70% of total production, and is supplemented by our best estimates of non-AMS orchards. Given these total tree numbers, average yields per tree are needed to complete the long-term forecasts. Yields from regional variety trials were initially used, but were found to be consistently higher than the average yields that growers were obtaining. Hence, a statistical model was developed using growers' historical yields, also taken from the AMS database. This model accounted for the effects of tree age, variety, year, region and tree spacing, and explained 65% of the total variation in the yield per tree data. The second level of crop prediction is an annual climate adjustment of these overall long-term estimates, taking into account the expected effects on production of the previous year's climate. This adjustment is based on relative historical yields, measured as the percentage deviance between expected and actual production. The dominant climatic variables are observed temperature, evaporation, solar radiation and modelled water stress. Initially, a number of alternate statistical models showed good agreement within the historical data, with jack-knife cross-validation R2 values of 96% or better. However, forecasts varied quite widely between these alternate models. Exploratory multivariate analyses and nearest-neighbour methods were used to investigate these differences. For 2001-2003, the overall forecasts were in the right direction (when compared with the long-term expected values), but were over-estimates. In 2004 the forecast was well under the observed production, and in 2005 the revised models produced a forecast within 5.1% of the actual production. Over the first five years of forecasting, the absolute deviance for the climate-adjustment models averaged 10.1%, just outside the targeted objective of 10%.

Achievements in forest tree genetic improvement in Australia and New Zealand 2: Development of Corymbia species and hybrids for plantations in eastern Australia

This paper describes the establishment of provenance seedling seed orchards of three spotted gums and cadaga (all species of Corymbia ex Eucalyptus). It also discusses the limitations of growing the spotted gums as pure species including: lack of mass flowering, susceptibility to a fungal shoot blight and low amenability to vegetative propagation. These limitations, together with observation of putative natural hybrids of the spotted gums with cadaga, and the early promise of manipulated hybrids, led to an intensive breeding and testing program. Many hybrid families have significant advantages in growth and tolerance to disease, insects and frost, and can be vegetatively propagated. They also exhibit broad environmental plasticity, allowing the best varieties to be planted across a wider range of sites than the spotted gums, resulting in more land being suitable for plantation development.

Screening selected myrtaceous tree species for production of essential oils in northern Queensland

Trials to identify alternative cropping options to Melaleuca alternifolia for northern Queensland essential oil growers were established at Dimbulah and Innot Hot Springs in 2001. Seed sources of Asteromyrtus symphyocarpa (1,8-cineole form), Eucalyptus staigeriana (citral), Melaleuca cajuputi subsp. cajuputi (trans-nerolidol), M. ericifolia (d-linalool), M. quinquenervia (trans-nerolidol and viridiflorol forms) and M. viridiflora (methyl cinnamate) with potential to produce commercial foliar oils were evaluated. Information was gathered on their adaptability, growth and oil yields over 49 months and 52 months (two harvests) from planting at Dimbulah and Innot Hot Springs, respectively. Of the species and chemotypes evaluated, M. quinquenervia showed potential for commercial production of trans-nerolidol, a compound used in perfumery. It had a very high survival rate (96%) and yields could be expected to improve dramatically from the average 100 kg/ha per harvest achieved in these trials with further research into selection of seed source, control of insect damage and breeding for genetic improvement. M. cajuputi subsp. cajuputi gave a similar performance to M. quinquenervia. The rarity of the trans-nerolidol form of this species and remoteness of its natural occurrence are impediments to further planting and research. E. staigeriana, with second harvest yields of ~600 kg/ha, performed exceptionally well on both sites but potential for development is limited by the ready availability of competitively priced E. staigeriana oil produced in South America. Survival of M. ericifolia ranged from 62% to 82% at 32 months (second harvest) at Innot Hot Springs and was deemed a failure at Dimbulah with poor growth and low survival, raising a major question about the suitability of this species for cultivation in the seasonally dry tropics. Planting of this species on a wider scale in northern Queensland cannot be recommended until more is known about factors affecting its survival. A. symphyocarpa and M. viridiflora were too slow-growing to warrant further consideration as potential oil-producing species at this time.

Comparison of decision tree, support vector machines, and Bayesian network approaches for classification of falls in Parkinson’s disease

Being able to accurately predict the risk of falling is crucial in patients with Parkinson’s dis- ease (PD). This is due to the unfavorable effect of falls, which can lower the quality of life as well as directly impact on survival. Three methods considered for predicting falls are decision trees (DT), Bayesian networks (BN), and support vector machines (SVM). Data on a 1-year prospective study conducted at IHBI, Australia, for 51 people with PD are used. Data processing are conducted using rpart and e1071 packages in R for DT and SVM, con- secutively; and Bayes Server 5.5 for the BN. The results show that BN and SVM produce consistently higher accuracy over the 12 months evaluation time points (average sensitivity and specificity > 92%) than DT (average sensitivity 88%, average specificity 72%). DT is prone to imbalanced data so needs to adjust for the misclassification cost. However, DT provides a straightforward, interpretable result and thus is appealing for helping to identify important items related to falls and to generate fallers’ profiles.

An (O)over-tilde(mn) Gomory-Hu Tree Construction Algorithm for Unweighted Graphs

We present a fast algorithm for computing a Gomory-Hu tree or cut tree for an unweighted undirected graph G = (V, E). The expected running time of our algorithm is (O) over tilde (mc) where vertical bar E vertical bar = m and c is the maximum u-v edge connectivity, where u, v is an element of V. When the input graph is also simple (i.e., it has no parallel edges), then the u-v edge connectivity for each pair of vertices u and v is at most n - 1; so the expected run-ning time of our algorithm for simple unweighted graphs is (O) over tilde (mn). All the algorithms currently known for constructing a Gomory-Hu tree [8, 9] use n - 1 minimum s-t cut (i.e., max flow) subroutines. This in conjunction with the current fastest (O) over tilde (n(20/9)) max flow algorithm due to Karger and Levine[11] yields the current best running time of (O) over tilde (n(20/9)n) for Gomory-Hu tree construction on simple unweighted graphs with m edges and n vertices. Thus we present the first (O) over tilde (mn) algorithm for constructing a Gomory-Hu tree for simple unweighted graphs. We do not use a max flow subroutine here; we present an efficient tree packing algorithm for computing Steiner edge connectivity and use this algorithm as our main subroutine. The advantage in using a tree packing algorithm for constructing a Gomory-Hu tree is that the work done in computing a minimum Steiner cut for a Steiner set S subset of V can be reused for computing a minimum Steiner cut for certain Steiner sets S' subset of S.