			Home About us Contact

Pattern Classification (pattern + classification)

Distribution by Scientific Domains

Medical Sciences	50%

Selected Abstracts

Semantic patterns for user-interactive question answering

CONCURRENCY AND COMPUTATION: PRACTICE & EXPERIENCE, Issue 7 2008
Tianyong Hao
Abstract A new type of semantic pattern is proposed in this paper, which can be used by users to post questions and answers in user-interactive question answering (QA) systems. The necessary procedures of using semantic patterns in a QA system are also presented, which include question structure analysis, pattern matching, pattern generation, pattern classification and answer extraction. Both the manual creation method and the automatic generation method are proposed for patterns for different applications. A pattern instantiation level metrics is also presented for the predication of the quality of generated or learned patterns. We implemented a user interface for using the semantic pattern in our QA system, which allows users to effectively post and answer questions. Copyright © 2007 John Wiley & Sons, Ltd. [source]

Diagnosis of invasion depth in early colorectal carcinoma by pit pattern analysis with magnifying endoscopy

DIGESTIVE ENDOSCOPY, Issue 2001
Shinji Tanaka
Background: The aim of this study was to clarify whether various pit patterns on the surface of colorectal tumors are associated with various levels of submucosal invasion. Methods: We examined pathologic features of the pit pattern of the tumor surface in 457 colorectal adenomas and early carcinomas. The examinations involved the use of magnifying endoscopy with indigocarmine dye spraying or crystal violet staining methods. Regarding the pit pattern classification, we used the types I, II, IIIL, IIIS, IV, VA and VN. We subclassified the VN pit pattern according to the area of the tumor surface covered into grades A (small), B (medium) and C (large). Results: Magnifying colonoscopic observation revealed the rates of submucosal invasion associated with specific pit patterns to be 1% (3/213) for IIIL, 5% (2/42) for IIIS, 8% (4/57) for IV, 14% (13/93) for VA and 80% (42/52) for VN. The rates of submucosal massive invasion (> 400 ,m) associated with specific pit patterns was 0% (0/213) for IIIL, 0% (0/42) for IIIS, 4% (2/57) for IV, 5% (5/93) for VA and 72% (38/52) for VN. Within the VN pit pattern subclassification, the incidence of submucosal invasion , 1500 ,m was found each grade (A, B & C): 5% (1/19) for grade A, 64% (14/22) for grade B and 93% (13/14) for grade C. Conclusion: Determination of pit pattern is useful for prediction of submucosal invasion depth and for decisions concerning treatment in colorectal tumors. Lesions with VA and non-grade C VN pit patterns are candidates for total endoscopic resection. A grade C VN pit pattern is a definite indicator of severely invasive submucosal carcinoma, which is unresectable by endoscopic resection. [source]

Intrapartum management guidelines based on fetal heart rate pattern classification

JOURNAL OF OBSTETRICS AND GYNAECOLOGY RESEARCH (ELECTRONIC), Issue 5 2010
Takashi Okai
First page of article [source]

Matching of catalogues by probabilistic pattern classification

MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY, Issue 1 2006
D. J. Rohde
ABSTRACT We consider the statistical problem of catalogue matching from a machine learning perspective with the goal of producing probabilistic outputs, and using all available information. A framework is provided that unifies two existing approaches to producing probabilistic outputs in the literature, one based on combining distribution estimates and the other based on combining probabilistic classifiers. We apply both of these to the problem of matching the H i Parkes All Sky Survey radio catalogue with large positional uncertainties to the much denser SuperCOSMOS catalogue with much smaller positional uncertainties. We demonstrate the utility of probabilistic outputs by a controllable completeness and efficiency trade-off and by identifying objects that have high probability of being rare. Finally, possible biasing effects in the output of these classifiers are also highlighted and discussed. [source]