Abstract
Cyber-Physical System (CPS) applications including human-robot interaction call for automated reasoning for rational decision-making. In the latter context, typically, audio-visual signals are employed. Τhis work considers brain signals for emotion recognition towards an effective human-robot interaction. An ElectroEncephaloGraphy (EEG) signal here is represented by an Intervals’ Number (IN). An IN-based, optimizable parametric k Nearest Neighbor (kNN) classifier scheme for decision-making by fuzzy lattice reasoning (FLR) is proposed, where the conventional distance between two points is replaced by a fuzzy order function (σ) for reasoning-by-analogy. A main advantage of the employment of INs is that no ad hoc feature extraction is required since an IN may represent all-order data statistics, the latter are the features considered implicitly. Four different fuzzy order functions are employed in this work. Experimental results demonstrate comparably the good performance of the proposed techniques.
Highlights
Cyber-Physical Systems (CPSs) have been defined as an integration of technical devices with both sensing and reasoning towards achieving an autonomy [1,2]
The work in [4] has proposed the engagement of Lattice Implication Algebra (LIA), that is the basis of Lattice-Valued Logic (LVL), in Fuzzy Lattice Reasoning (FLR) based on a lattice of real number intervals
Lattice Computing (LC) models are promising in CPS applications especially regarding human-robot interaction since they can (i) process numerical as well as non-numerical data towards computing with semantics represented by a partial-order relation, (ii) cope with uncertainties represented by information granules, (iii) naturally engage logic and/or reasoning, and (iv) process data fast [28]
Summary
Cyber-Physical Systems (CPSs) have been defined as an integration of technical devices with both sensing and reasoning towards achieving an autonomy [1,2]. LC models are promising in CPS applications especially regarding human-robot interaction since they can (i) process numerical as well as non-numerical data towards computing with semantics represented by a partial-order relation, (ii) cope with uncertainties represented by information granules, (iii) naturally engage logic and/or reasoning, and (iv) process data fast [28]. The results by four different fuzzy order functions (σ) are demonstrated here comparatively The latter approach has not been applied in EEG emotion recognition previously by any other method. A parametric k Nearest Neighbor (kNN) involving INs is introduced here for emotion classification, where a metric distance (d) has been replaced by a fuzzy order function (σ) for reasoning-by-analogy Another novelty of this work is the employment of a recently introduced IN induction technique, which suggests an improved IN interpretation [1]. Tables of abbreviations (Table A1), nomenclatures (Table A2), and symbols (Table A3) are provided in the Appendix A
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
