Tobi Knaup

This work focuses on automatically analyzing a speaker's sentiment in online videos containing movie reviews. In addition to textual information, this approach considers adding audio features as typically used in speech-based emotion recognition as well as video features encoding valuable valence information conveyed by the speaker. Experimental results indicate that training on written movie reviews is a promising alternative to exclusively using (spoken) in-domain data for building a system… Show more

This work focuses on automatically analyzing a speaker's sentiment in online videos containing movie reviews. In addition to textual information, this approach considers adding audio features as typically used in speech-based emotion recognition as well as video features encoding valuable valence information conveyed by the speaker. Experimental results indicate that training on written movie reviews is a promising alternative to exclusively using (spoken) in-domain data for building a system that analyzes spoken movie review videos, and that language-independent audio-visual analysis can compete with linguistic analysis. Show less

We propose a data-driven approach based on back-off N-Grams and Support Vector Machines, which have recently become popular in the fields of sentiment and emotion recognition. In addition, we introduce a novel valence classifier based on linguistic analysis and the on-line knowledge sources ConceptNet, General Inquirer, and WordNet. As special benefit, this approach does not demand labeled training data. Moreover, we show how such knowledge sources can be leveraged to reduce out-of-vocabulary… Show more

We propose a data-driven approach based on back-off N-Grams and Support Vector Machines, which have recently become popular in the fields of sentiment and emotion recognition. In addition, we introduce a novel valence classifier based on linguistic analysis and the on-line knowledge sources ConceptNet, General Inquirer, and WordNet. As special benefit, this approach does not demand labeled training data. Moreover, we show how such knowledge sources can be leveraged to reduce out-of-vocabulary events in learning-based processing. To profit from both of the two generally different concepts and independent knowledge sources, we employ information fusion techniques to combine their strengths, which ultimately leads to better overall performance. Finally, we extend the data-driven classifier to solve a regression problem in order to obtain a more fine-grained resolution of valence. Show less

In the fields of sentiment and emotion recognition, bag of words modeling has lately become popular for the estimation of valence in text. A typical application is the evaluation of reviews of e. g. movies, music, or games. In this respect we suggest the use of back-off N-Grams as basis for a vector space construction in order to combine advantages of word-order modeling and easy integration into potential acoustic feature vectors intended for spoken document retrieval. For a fine granular… Show more

In the fields of sentiment and emotion recognition, bag of words modeling has lately become popular for the estimation of valence in text. A typical application is the evaluation of reviews of e. g. movies, music, or games. In this respect we suggest the use of back-off N-Grams as basis for a vector space construction in order to combine advantages of word-order modeling and easy integration into potential acoustic feature vectors intended for spoken document retrieval. For a fine granular estimate we consider data-driven regression next to classification based on Support Vector Machines. Alternatively the on-line knowledge sources ConceptNet, General Inquirer, and WordNet not only serve to reduce out-of-vocabulary events, but also as basis for a purely linguistic analysis. As special benefit, this approach does not demand labeled training data. A large set of 100 k movie reviews of 20 years stemming from Metacritic is utilized throughout extensive parameter discussion and comparative evaluation effectively demonstrating efficiency of the proposed methods. Show less