RESUMO
To comprehend the genuine reading habits and preferences of diverse user cohorts and furnish tailored reading recommendations, this study introduces an English text reading recommendation model designed specifically for long-tail users. This model integrates collaborative filtering algorithms with the FastText classification method. Initially, the integrated collaborative filtering algorithm is explicated, followed by the calculation of the user's interest distribution across various types of English texts, achieved through an enhanced Ebbinghaus forgetting curve and analysis of user reading behaviors. Subsequently, an intelligent English text reading recommendation is generated by amalgamating collaborative filtering algorithms with association rule-based recommendation algorithms. Through optimization of the recommendation generation process, the model's recommendation accuracy is enhanced, thereby augmenting the performance and user satisfaction of the recommendation system. Finally, a comparative analysis is conducted with respect to the Top-N algorithm model, matrix factorization-based algorithm model, and FastText classification model, illustrating the superior recommendation accuracy and F-Measure value of the proposed model. The study findings indicate that when the recommendation list contains 10, 30, 50, and 70 texts, the recommendation accuracy of the proposed algorithm model is 0.75, 0.79, 0.8, and 0.74, respectively, outperforming other algorithms. Furthermore, as the number of texts increases, the F-Measure of all four models gradually improves, with the final F-Measure of the proposed model reaching 0.81. Notably, the F-Measure of the English text reading recommendation model proposed in this study significantly surpasses that of the other three recommendation methods. Demonstrating commendable performance in recall rate, root mean square error, normalized cumulative gain, precision, and accuracy, the model adeptly reflects user reading interests, thereby enhancing the accuracy of text recommendations and the overall system performance. The study findings offer crucial insights and guidance for enhancing the accuracy and overall efficacy of English text recommendation systems.
RESUMO
The critical success index (CSI) is an established metric used in meteorology to verify the accuracy of weather forecasts. It is defined as the ratio of hits to the sum of hits, false alarms, and misses. Translationally, CSI has gained popularity as a unitary outcome measure in various clinical situations where large numbers of true negatives may influence the interpretation of other, more traditional, outcome measures, such as specificity (Spec) and negative predictive value (NPV), or when unified interpretation of positive predictive value (PPV) and sensitivity (Sens) is needed. The derivation of CSI from measures including PPV has prompted questions as to whether and how CSI values may vary with disease prevalence (P), just as PPV estimates are dependent on P, and hence whether CSI values are generalizable between studies with differing prevalences. As no detailed study of the relation of CSI to prevalence has been undertaken hitherto, the dataset of a previously published test accuracy study of a cognitive screening instrument was interrogated to address this question. Three different methods were used to examine the change in CSI across a range of prevalences, using both the Bayes formula and equations directly relating CSI to Sens, PPV, P, and the test threshold (Q). These approaches showed that, as expected, CSI does vary with prevalence, but the dependence differs according to the method of calculation that is adopted. Bayesian rescaling of both Sens and PPV generates a concave curve, suggesting that CSI will be maximal at a particular prevalence, which may vary according to the particular dataset.