<p>Results are mean (standard derivation) obtained by 10-fold cross-validation experiments on MovieLens (5,000 users and 5,977 objects) with cosine similarity measure. Restart probabilities for random walk approaches are set to 0.9. <i>MRR</i> represents mean relative rank, <i>PR@20</i> represents precision at the default <i>L</i> value of 20, <i>RE</i> represents recall enhancement, <i>HR@20</i> represents hit-rate at <i>L</i> = 20, <i>MP</i> represents mean personalization, <i>MN</i> represents mean novelty.</p><p>Performance of different methods.</p
<p>Sn: sensitivity (recall).</p><p>Pr: precision.</p><p>Sp: specificity.</p><p>Ac: accuracy.</p><p>M...
<p>AR: Accuracy rate, SE: Sensitivity, SP: Specificity, PPV: Positive predictive value, NPV: Negativ...
<p>Number of recommendation is 100. There was no significant change in the performance rank among me...
<p>Results are obtained by 10-fold cross-validation experiments on MovieLens (5,000 users and 5,977 ...
<p>(A) Mean relative rank. (B) Precision at <i>L</i> = 20. (C) Recall enhancement. (D) Hit-rate at <...
<p>Results are obtained by 10-fold cross-validation experiments on MovieLens (5,000 users and 5,977 ...
<p>(A) Proportions of test objects ranked at top positions. (B) Cumulative distributions of top rank...
<p>For the MLR, ANN and RF methods, 95% confidence intervals of the difference between the indicator...
<p>The values are given for many choices of the reference test and comparison modalities and various...
<p>(A–C) Recall enhancement. (D–F) Hit-rate at <i>L</i> = 20. Results are obtained by 10-fold cross-...
<p><i>F: F</i>-score; <i>P:</i> precision; <i>R:</i> recall; <i>Acc:</i> accuracy; <i>MCC:</i> Matth...
<p>Shown is the mean performance for each machine learning method (see <a href="http://www.plosone.o...
Developing state-of-the-art approaches for specific tasks is a major driving force in our research c...
<p>Average ranks of four different types of ensembles with 10 constituents combined by averaging. Th...
<p>Results show the mean accuracy (upper plot) and proportion of significant results (bottom plot) o...
<p>Sn: sensitivity (recall).</p><p>Pr: precision.</p><p>Sp: specificity.</p><p>Ac: accuracy.</p><p>M...
<p>AR: Accuracy rate, SE: Sensitivity, SP: Specificity, PPV: Positive predictive value, NPV: Negativ...
<p>Number of recommendation is 100. There was no significant change in the performance rank among me...
<p>Results are obtained by 10-fold cross-validation experiments on MovieLens (5,000 users and 5,977 ...
<p>(A) Mean relative rank. (B) Precision at <i>L</i> = 20. (C) Recall enhancement. (D) Hit-rate at <...
<p>Results are obtained by 10-fold cross-validation experiments on MovieLens (5,000 users and 5,977 ...
<p>(A) Proportions of test objects ranked at top positions. (B) Cumulative distributions of top rank...
<p>For the MLR, ANN and RF methods, 95% confidence intervals of the difference between the indicator...
<p>The values are given for many choices of the reference test and comparison modalities and various...
<p>(A–C) Recall enhancement. (D–F) Hit-rate at <i>L</i> = 20. Results are obtained by 10-fold cross-...
<p><i>F: F</i>-score; <i>P:</i> precision; <i>R:</i> recall; <i>Acc:</i> accuracy; <i>MCC:</i> Matth...
<p>Shown is the mean performance for each machine learning method (see <a href="http://www.plosone.o...
Developing state-of-the-art approaches for specific tasks is a major driving force in our research c...
<p>Average ranks of four different types of ensembles with 10 constituents combined by averaging. Th...
<p>Results show the mean accuracy (upper plot) and proportion of significant results (bottom plot) o...
<p>Sn: sensitivity (recall).</p><p>Pr: precision.</p><p>Sp: specificity.</p><p>Ac: accuracy.</p><p>M...
<p>AR: Accuracy rate, SE: Sensitivity, SP: Specificity, PPV: Positive predictive value, NPV: Negativ...
<p>Number of recommendation is 100. There was no significant change in the performance rank among me...