... structures (e.g. linear-chain), anexact inference can be obtained efficiently if the num-ber of output labels is not large. However, for largenumber of output labels, the inference is often pro-hibitively ... {yt}Tt=1are sequences of in-put and output variables. Respectively, where T is thelength of sequence, xt∈ X and yt∈ Y where X is thefinite set of the input observations and Y is that of theoutput ... becauseour goal is to evaluate the efficiency of inference algo-rithms. Moreover, using the previous state -of- the-artfeatures we expect the achievement of better accuracy.All our models were trained...