Share this post on:

Her categories in our study. SCLIWC specificity was larger than 0.70 for many of those categories except constructive emotion (.69) and discrepancy (.67). In Study 1, through correlation analysis in between human ratings and SCLIWC variables of word count as those did in earlier research [7,27], we identified substantial correlations with the coefficients of generally the identical level in those studies. If making use of status count scoring strategy in calculating SCLIWC variables on Weibo statuses, the correlation coefficients may very well be even higher. In Study 2, the evaluating primarily based on signal-detection theory showed high sensitivity of SCLIWC for various categories and higher specificity as well as adverse predictive value for many categories. The validities on distinctive LIWC categories had been of large differences, as in Pennebaker et al.’s reports [28]. The difference on both content and style of your three Net text varieties had been also reflected in our final results, in both the word percentage profiles also because the correlations among human ratings and SCLIWC variables. The discrepancies of word percentage profiles have been consistent with our common sense about these text sorts, even though the discrepancies from the correlation coefficients may possibly present some strategies in application, that there exists some distinction of SCLIWC validity on various varieties of Internet text. As an example, the accuracies of detecting expressions of affective processes employing SCLIWC on Weibo statuses, Renren blogs and news comments were quite close, whilst the Protodioscin price lexicon seemed to execute better in detecting concerned contents in Renren blogs and news comments.How to PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/21187425 use SCLIWC to detect psychological expression in SNS brief internet texts far more efficientlyOur second question, that ways to use SCLIWC in a more effective approach to detect psychological expressions on SNS brief text, was explored in two elements: the quantity of texts as well as the scoring system. The earlier studies relating to LIWC validity usually used written supplies with aPLOS A single | DOI:ten.1371/journal.pone.0157947 June 20,ten /Evaluating the Validity of Simplified Chinese LIWCsubstantial volume of words, for example essays of private writing [7] or messages in onlinebased support groups [27,29]. While it may not be formally discussed just before, it really is uncomplicated to understand that the instruments primarily based on lexicons, for example LIWC, could execute improved when the words inside the material reach or exceed a particular quantity. We located that SCLIWC could effectively detect psychological expressions (medium to high correlations with human ratings) of numerous categories, even when the material limited to one particular day’s Weibo statuses (about 75?00 words on typical), and because the level of statuses increasing to a week’s along with a month’s, SCLIWC could be valid on far more categories. In addition to counting words, the usual way of processing texts with LIWC, counting statuses could possibly be one more selection when processing a set of Weibo statuses. As in our outcomes, the status count score typically had greater correlations with human ratings on a day/week/month’s Weibo statuses, than the word count score had. In particular for the results on a month’s statuses, the correlations among status count score and human ratings have been considerable on all chosen categories, which indicated that the status count scoring approach was a pretty efficient processing method while working with SCLIWC to detect psychological expressions from a set of SNS brief texts. There is certainly a crucial difference amongst a set of SNS quick texts and most supplies pro.

Share this post on:

Author: bet-bromodomain.