Evaluating the Research in Automatic Emotion Recognition

Aims: In this paper, we present a systematic literature review of the scientific research on the automatic recognition of personal emotions. Based on an overview of the existing evidence, we further discuss the main limitations, future directions, institutions, and influential figures in this field. Method: We conducted a systematic literature search of four electronic databases. Papers were systematically analysed in relation to specific research questions. Results: The literature review yielded a pool of 110 references, which was narrowed down to a final set of 32 based on inclusion and exclusion criteria. Findings reveal a growing volume of research in this field, with the USA and the UK leading these efforts and the researcher Maja Pantic as a pioneer and central figure in this area. Overall, our work revealed nine topics, with 98 databases identified. Further examination of the selected works exposed specific problems. Conclusions: Despite the growing number of existing research in each area of this vast field, most of the work is not systematic. To increase the efficiency and efficacy of obtaining information, we recommend the use of the standard systematic literature method in future efforts toward information syntheses on each of the topics.