Utility of real-time decision-making in commercial data stream mining domains

The objective is to measure utility of real-time commercial decision making. It is important due to a higher possibility of mistakes in real-time decisions, problems with recording actual occurrences, and significant costs associated with predictions produced by algorithms. The first contribution is to use overall utility and represent individual utility with a monetary value instead of a prediction. The second is to calculate the benefit from predictions using the utility-based decision threshold. The third is to incorporate cost of predictions. For experiments, overall utility is used to evaluate communal and spike detection, and their adaptive versions. The overall utility results show that with fewer alerts, communal detection is better than spike detection. With more alerts, adaptive communal and spike detection are better than their static versions. To maximise overall utility with all algorithms, only 1% to 4% in the highest predictions should be alerts.

Riverine macroinvertebrate assemblages up to 8 years

Willow removal followed by riparian revegetation is a widespread river restoration practice in Australia, but the ecological response to this has rarely been evaluated. We sampled river macroinvertebrates from six sites each of three riparian vegetation types: revegetated (treatment), willow-dominated (control) and native forest (reference) in the Gellibrand River catchment during austral spring 2007 and autumn 2008, and measured temperature and light intensity. Revegetated sites varied in age from 1 to 8 years since restoration. Abundances of invertebrates were similar across vegetation types, but were higher during autumn. Macroinvertebrate assemblages at revegetated sites (regardless of age) and at willow-dominated sites showed little among-site variation compared with native forest sites,which showed high site-to-site variability. Water temperatures and light intensity were higher at revegetated sites where works had recently been completed and cooler in native forest sites and long-established revegetated sites. The reduced variability in macroinvertebrate communities among revegetated sites may result from their history as willow-dominated sites or from the disturbance created by willow removal. Either way, these results suggest that longer than 8 years is required before macroinvertebrate assemblages in restored stretches of stream show the variation that appears characteristic of natural sites.

Resource heterogeneity for stream detritivores

Invertebrate animals in headwater streams depend upon leaves, wood and bark for both food and living space. This project demonstrated the strength of that dependence where the patchiness in habitat is influenced at the smallest spatial scales by the movement of sediments, and at the largest by agricultural clearing.

Stream bryophytes of Victorian temperate rainforests

Cool Temperate, Warm Temperate and Gallery Rainforests of Victoria were examined and found to heave distinct suites of stream bryophytes, although a ubiquitous group occurred. Bryophytes represented seven biogeographic elements; most were Australasian. Species reproducing sexually did so synchronously but most species reproduced asexually.

Incremental and adaptive clustering stream data over sliding window

Cluster analysis has played a key role in data stream understanding. The problem is difficult when the clustering task is considered in a sliding window model in which the requirement of outdated data elimination must be dealt with properly. We propose SWEM algorithm that is designed based on the Expectation Maximization technique to address these challenges. Equipped in SWEM is the capability to compute clusters incrementally using a small number of statistics summarized over the stream and the capability to adapt to the stream distribution’s changes. The feasibility of SWEM has been verified via a number of experiments and we show that it is superior than Clustream algorithm, for both synthetic and real datasets.

Restoration of the habitat-forming fucoid alga Hormosira banksii at effluent-affected sites : competitive exclusion by coralline turfs

Fucoid algae often dominate intertidal rocky shores, providing habitat and modifying ecosystem resources for other species, but are susceptible to discharge of sewage effluent. In this study we assessed the potential for competition from coralline turfs to inhibit restoration of the intertidal fucoid macroalga Hormosira banksii at sites associated with an ocean outfall a scenario of improving following water quality in the nearshore coastal environment. The percentage cover and number of individuals of H. banksii were negatively correlated with both the percentage cover and turf height of Corallina officinalis. In contrast, H. banksii was positively associated with rocky substrata and recruited well to rock-surface substrata. Importantly, there appears to be a threshold abundance where the percentage cover of H. banksii rarely reaches above 20% cover amongst coralline turfs with >40% cover. These data support a model of alternative community states: H. banksii dominated canopy on rocky substrata versus C. officinalis turf. In field and laboratory experiments, extensive coralline turfs (up to 4 cm thick) were shown to inhibit recruitment of H. banksii. This study shows competitive exclusion by coralline turfs may limit the successful restoration of habitat provided by H. banksii to shores that have been affected by sustained discharge of secondarily treated sewage effluent. We suggest potential strategies for management authorities to consider when seeking ways of restoring fucoid communities affected by anthropogenic disturbances such as wastewater disposal.