Restricted evaluation in information retrieval

Bollmann, P.; Cherniavsky, V. S.

doi:10.1145/1013228.511757

Cited by 10 publications

(16 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These are defined in Heine (1984). We note also that the MZ-metric is monotone with van Rijsbergen's E measure (Van Rijsbergen, 1974), and the criticalevaluative comments on E and Retrieval Power by Bollmann (1984) and Bollman and Cherniavsky (1981).…”

Section: Measures Of Retrieval Effectiveness and Their Relevance To mentioning

confidence: 69%

An investigation of the optimization of search logic for the MEDLINE database

Heine

Tague

1991

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

One of the key variables in the optimization of the retrieval process is the logic imposed on a set of query terms. If the query is small, a combinatorial algorithm can be employed to identify search expressions having an optimal logical form. An experiment is described in which this was done for queries expressed against MEDLINE, for a variety of criterion variables. The method employed is useful not only for assisting in identifying optimal logical forms, as demonstrated, but also as an experimental control device to assist investigations into the effects of varying the set, and number, of search terms. The experiment also suggests several novel properties of effective searching against MEDLINE, for example that searching with four MeSH terms is likely to be more successful than searching with a lesser number of terms, provided search logic is optimal. The latter result is, surprisingly, true for both Precision and Recall, i.e., a tradeoff between the maximally attainable values of these variables fails to hold when the number of search terms varies in this range.

show abstract

Section: Measures Of Retrieval Effectiveness and Their Relevance To mentioning

confidence: 69%

An investigation of the optimization of search logic for the MEDLINE database

Heine

Tague

1991

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…We set = 0.0000001 producing minimum test errors for selecting features. Finally, we employ precision, recall and F-measure (Bollmann and Cherniavsky, 1981), widely used measure to evaluate the performance of all the methods mentioned above Yang et al, 2014). The Fmeasure is the harmonic mean of precision (p) and recall (r), which is represented as 2×p×r/(p+r).…”

Section: Experimental Data Settings and Evaluation Metricsmentioning

confidence: 99%

“…The orphan nuclear receptors, ERRalpha and gamma affect cardiac functions (Dufour et al, 2007). Regarding GO 0001906 (cell killing), both gradual and acute cell death is hallmarks of cardiac pathology, including heart failure, myocardial infarction and ischemia/reperfusion (Bollmann and Cherniavsky, 1981).…”

Section: Feature Analysismentioning

confidence: 99%

Biological Feature Selection and Disease Gene Identification using New Stepwise Random Forests

Hwang¹

2017

Industrial Engineering and Management Systems

View full text Add to dashboard Cite

Identifying disease genes from human genome is a critical task in biomedical research. Important biological features to distinguish the disease genes from the non-disease genes have been mainly selected based on traditional feature selection approaches. However, the traditional feature selection approaches unnecessarily consider many unimportant biological features. As a result, although some of the existing classification techniques have been applied to disease gene identification, the prediction performance was not satisfactory. A small set of the most important biological features can enhance the accuracy of disease gene identification, as well as provide potentially useful knowledge for biologists or clinicians, who can further investigate the selected biological features as well as the potential disease genes. In this paper, we propose a new stepwise random forests (SRF) approach for biological feature selection and disease gene identification. The SRF approach consists of two stages. In the first stage, only important biological features are iteratively selected in a forward selection manner based on one-dimensional random forest regression, where the updated residual vector is considered as the current response vector. We can then determine a small set of important biological features. In the second stage, random forests classification with regard to the selected biological features is applied to identify disease genes. Our extensive experiments show that the proposed SRF approach outperforms the existing feature selection and classification techniques in terms of biological feature selection and disease gene identification.

show abstract

“…In order to evaluate and compare these supervised learning techniques for disease gene prediction, we used 10-fold cross-validation on the training set and measured the performance by F-measure [49]. This measure is harmonic mean of precision and recall, so it reflects an average effect of both precision and recall.…”

Section: Classification Techniques Parameter Settings and Validationmentioning

confidence: 99%

A Comparative Study of Classification-Based Machine Learning Methods for Novel Disease Gene Prediction

Hoài²,

Kwon

2015

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Abstract. Prediction of novel genes associated to a disease is an important issue in biomedical research. At early days, annotation-based methods were proposed for this problem. In next stage, with high-throughput technologies, data of interaction between genes/proteins has grown quickly and covered almost genome and proteome, and therefore network-based methods for the issue is becoming prominent. Besides those two methods, the prediction problem can be also approached using machine learning techniques because it can be formulated as a classification task of machine learning. To date, a number of supervised learning techniques and various types of gene/protein annotation data have been used to solve the disease gene classification/prediction problem. However, to the best of our knowledge, there has been no study on the comparison of these methods that work on comprehensive biomedical annotation data. In addition, it is generally true that no classifier is better than others for all classification problems. Therefore, in this study, we compare the performance of disease gene prediction of several supervised learning techniques that have been used in the literature such as Decision Tree Learning, k-Nearest Neighbor, Naive Bayesian, Artificial Neural Networks and Support Vector Machines. We additionally assess Random Forest, a relatively new decision-tree-based ensemble learning method. The simulation results indicate that Random Forest obtained the best performance of all. Also, all methods are stable with the change of known disease genes used as positive training samples.

show abstract

Restricted evaluation in information retrieval

Cited by 10 publications

References 6 publications

An investigation of the optimization of search logic for the MEDLINE database

An investigation of the optimization of search logic for the MEDLINE database

Biological Feature Selection and Disease Gene Identification using New Stepwise Random Forests

A Comparative Study of Classification-Based Machine Learning Methods for Novel Disease Gene Prediction

Contact Info

Product

Resources

About