This article surveys probablistic approaches to modeling information retrieval. The basic concepts of probabilistic approaches to information retrieval are outlined and the principles and assumptions upon which the approaches are based are presented. The various models proposed in the development of IR are described, classified, and compared using a common formalism. New approaches that constitute the basis of future research are described.

Aerobic yeasts of the genera Pichia and Williopis are commonly regarded as spoilage yeasts of beer and wine by causing turbidity, a surface film of yeast growth and often an excessive estery flavour. Hozvever, their ability to utilise sugars oxidatively for cell growth with the production of estery and other flavours of wine with only minimal production ofethanol suggests a method for the production of low-alcohol ivines of pleasant "fermented" flavour without the need for additional equipment to remove alcohol by dialysis, reverse osmosis or distillation, or ivithout the excessive sweetness remaining from arrested fermentation.Three strains of Pichia and one of Williopsis were examined for their ability to produce approximately 3%(v/v) ethanol and a good estery and fruity flavour. With normal anaerobic fermentation conditions, or with gentle stirring to preventformation of a surface film, excessive amounts of alcohol were produced from grape juice of 15% or 20% (w/v) initial sugar concentration. However, an acceptably flavoured wine of alcohol content < 3% was produced by agitation and aeration during fermentation. The ethanol formed in the early stages of culture was oxidised to a final level < 3%, with the production of cell mass and an acceptable flavour.

No abstract

OBJECTIVE To prospectively compare the efficacy of bi‐basal vs bi‐apical periprostatic nerve block (PPNB) during 12‐core prostate biopsy guided by transrectal ultrasonography (TRUS), and to evaluate the pain experienced on inserting the probe compared to the biopsy procedure, as PPNB with lignocaine local anaesthesia has been used for over a decade for minimizing pain during prostatic biopsy. PATIENTS AND METHODS In all, 143 men who were to have a TRUS‐guided prostate biopsy were systematically randomized to two groups, to receive PPNB at the apex or base. A 10‐cm visual analogue score was used to record the pain experienced during probe insertion, the biopsy and just before to leaving the department . RESULTS The mean pain score on biopsy in the apical group was similar to that of the basal group (apex 1.9, base 1.6, P = 0.36). Probe introduction produced a significantly higher pain score (probe 2.2, biopsy 1.7, P < 0.001) than at the biopsy. CONCLUSIONS Patients who experienced greater pain with the introduction of the probe also reported more pain with the biopsy procedure. The site of local anaesthetic before prostatic biopsy showed no significant difference in pain scores. Older men tolerated the procedure better. Analgesia after PPNB at near either the apex or base appears equal, regardless of the site of injection. We suggest that topical perianal anaesthetic agents could significantly reduce not only pain perception, but also improve tolerance.

Phytoestrogenic substances have previously been isolated and identified in two alcoholic beverages: bourbon and beer. To delineate the relative potencies of the estrogenic substances of plant origin thus far identified in these commonly consumed alcoholic beverages, we evaluated the ability of biochanin A, beta-sitosterol, genistein, and daidzein to bind to cytosolic estrogen receptor binding sites. The in vitro studies demonstrated that each of the contained substances was capable of effectively competing for cytosolic estrogen receptor binding sites of rat liver and uterus. Further, the two phytoestrogenic constituents of bourbon, beta-sitosterol and biochanin A, were less potent than those present in beer. Given the high concentration of beta-sitosterol in bourbon, we chose to evaluate the estrogenicity of beta-sitosterol in vivo using ovariectomized rats. beta-sitosterol was administered either daily or intermittently at 3 doses, based on amounts previously determined to be present in bourbon. The in vivo studies demonstrated that beta-sitosterol is capable of producing a weak estrogenic effect only at the lowest dose (6.2 micrograms/dl) administered intermittently. These responses suggest that beta-sitosterol may be weakly estrogenic at low doses, but is unable to maintain such an effect at higher doses.