A method for automatically generating the emotional vectors of emoticons using weblog articles

This study proposes a method to create emotional vectors of emoticons automatically using the collocation relationship between emotional words and emoticons which is derived from many weblog articles.

Abstract

(1) Background: The isokinetic measurement (IM) of the leg muscles is well established but costly, whereas the Bunkie Test (BT) is a rarely investigated but easy-to-conduct functional test to evaluate the total posterior chain. Although the tests differ in aim and test structures, both have their justification in the assessment process. Therefore, this study evaluated the diagnostic accuracy of the BT and the IM. (2) Methods: 21 participants (9 female, 12 male; age, 26.2 ± 5.26 years; weight 73.8 ± 14.6 kg; height 176.0 ± 9.91 cm) and 21 patients (9 female, 12 male; age, 26.5 ± 5.56 years; weight, 72.6 ± 16.9 kg; height 177.0 ± 10.1 cm) with self-reported pain in the knee performed the IM and the BT. For IM, we calculated the ratio of the knee mean flexor/extensor peak torque (H/Q ratio) for 60°/s and 120°/s, and BT performance was measured in seconds. We classified the IM (p-value of ≤0.05 is considered significant. (3) Results: The sensitivity for the BT was 0.89, 95% CI [0.67, 0.99], and the specificity was 0.52 [0.30, 0.74]. For the IM, the sensitivity was 0.14 [0.03, 0.36] for 60°/s and 0.05 [0.00, 0.24] for 120°/s, and the specificity was 0.70 [0.46, 0.88] for 60°/s and 0.90 [0.68, 0.99] for 120°/s. The results of the Chi-Square tests were significant for the BT (χ2 (1) = 6.17, p = 0.01) but not for the IM (60°/s: χ2 (1) = 0.70, p = 0.40; 120°/s: χ2 (1) = 0.00, p = 0.97). (4) Conclusions: Patients were more likely to obtain a positive test result for the BT but not for the IM.