Introducing situational signs in qualitative probabilistic networks

CPD Tree Learning Using Contexts as Background Knowledge

“…Some authors suggest to tune a KB when conflicts have been detected [7]. In [8], the model is capable to evolve, depending of its current state. Contrary to these works, our paper assumes the existence of a well-formed KB of motifs and examines the impact of its incorporation into learning algorithms.…”

QMIQPN: An enhanced QPN based on qualitative mutual information for reducing ambiguity

“…Since the probability of superposition computation at most is the minimal range of formula (20) and (21), the probability range of variables X and Z taking the same values at least is the maximal range of formula (20) and (21), that is, b 6 Prðx; zÞ þ Prð x; zÞ 6 1; moreover, the composition of qualitative influences in a basic QPN,…”

“…Furthermore, reasoning with qualitative probabilities is much more efficient than reasoning with precise ones, the inference complexity of QPN is a polynomial in the size of the network [5], rather than NP-hard [6]. Therefore, many approaches have been proposed for QPN modeling and inference according to various kinds of applications [7][8][9][10][11][12][13][14][15][16][17][18][19][20].…”

Qualitative probabilistic networks with reduced ambiguities

“…On one hand, for the arc-related ambiguities in a QPN, Bolt et al extended the framework of QPNs with the concepts of situational signs, and a situational sign is associated with a non-monotonic influence and captures information about the effect of the influence in the current state of the network [1,2]. Renooij et al [26] extended the basic formalism of QPNs by providing for the inclusion of contextspecific information about influences and showed that exploiting this information upon inference had the ability to forestall unnecessarily weak results.…”