Query Results

Number of results 0
Abstracts of input examples inputs.html
Complete abstracts of the Medline records given as relevant training examples. They have been ranked by classifier scores.
PubMed IDs of results results.txt
PubMed IDs and scores of classifier predictions, ranked by decreasing score.
PubMed IDs of inputs inputs.txt
PubMed IDs and scores of the relevant training examples.
Started at 2009/03/25 12:59:45 GMT
Time at which query was started
Finished at 2009/03/25 13:04:28 GMT
Time at which this file was written.
Feature score method scores_laplace_split
Name of the method used to calculate feature scores. Docstring for the method: For feature probabilities we use a Laplace prior, of 1 success and 1 failure in total, split between the classes according to size. This avoids problems with class skew.
Min Information Gain 2e-05
We exclude features with less than this value of Information Gain.
Base score -68.0409832312
The log likelihood ratio of an empty article (one in which every feature failed to occur).
Prior score -12.0272975925
The log of the prior probability ratio for an article being relevant versus irrelevant (added to log likelihood ratio to obtain the final score). Equals the logit of the estimated prevalence of relevant articles in Medline (which may be estimated from the input size or specified separately).
Limit 1000
The maximum number of results to include.
Threshold 0.0
Default Naive Bayes classification threshold is zero. This threshold is the minimum log probability ratio for predicting an article to be relevant.
Minimum date 20071001
The minimum date considered when parsing Medline (both when making feature counts, and when querying)

Feature Statistics

Quantity Relevant Docs Irrelevant Docs
Number of documents 4 836293
Number of selected, occurring features 342 34885
Total occurrences of selected features 370 52805801
Selected features per Medline record 92.500 63.143
Of the considered feature types, 34892 features are selected out of 1465125 occurring at least once in training data. The aggressivity of selection is 41.990. The complete database lists 3703762 potential features.

Features with high TF.IDF

Features with TF.IDF above 0.2 or 0.3 could make good keywords. TF.IDF is term frequency times inverse document frequency, where we treat the set of input citations as a single document

TF-IDF Type Term Term ID Score Pos Neg
0.05 mesh Lymphangioma 52366 9.28 2 76
0.05 mesh Splenic Neoplasms 40839 8.67 2 141
0.04 a l ierardi-curto 2817087 11.84 1 0
0.04 a jn foss 2608027 11.84 1 0
0.04 a wa pryor 584366 11.84 1 0
0.04 issn 0271-5333 529036 11.84 1 0
0.04 a p chanmugam 405370 11.84 1 0
0.04 issn 0141-8955 171516 11.84 1 0
0.04 a o masamune 50609 11.84 1 0
0.03 w Splenic 4402 6.42 2 1354
0.03 a kh sohn 1118735 11.15 1 2
0.03 a m boudreau 489350 11.15 1 2
0.03 a s saitta 405778 11.15 1 2
0.03 w Pattern-oriented 1235128 10.93 1 3
0.03 a y hamashima 630926 10.93 1 3
0.03 w Cysts 5167 6.11 2 1844
0.03 a t komatsuda 1583598 10.75 1 4
0.03 a a paterson 577915 10.59 1 5
0.03 a gs bisset 571838 10.59 1 5
0.03 a gt berry 523304 10.59 1 5