This paper describes 'knowledge-based' data-mining techniques, developed for the assessment of landslide susceptibility and hazard with particular reference to its application in the Wollongong area. Large scale maps of geology and a comprehensive Landslide Inventory with regional coverage have been prepared. GIS-based derivatives of the digital elevation model including slope, geomorphology, curvature, flow accumulation and wetness index have been developed. Model performance has been assessed as part of a refined methodology for validation, including field inspections. Susceptibility zones outside known landslide areas have been classified as (a) high (b) moderate, (c) low and (d) very low susceptibility. Results show the high susceptibility zone covers 10% of the study area and contains 60% of known landslides, the moderate zone covers 12% of the study area and contains 32% of known landslides, the low zone covers 6.4% of area and contains 3.3% of known landslides and the very low zone covers 71 % of the study area and contains 4% of the landslides. The susceptibility maps have been upgraded to hazard level maps with identification of individual zone landslide likelihoods, specific landslide frequency, volume and 'profile' angles. The paper concludes with a preliminary landslide susceptibility map for a segment of the Sydney Basin Region developed using the methodology described in this paper.