In this paper, we investigate the problem of learning user preferences in the transductive setting. The key idea of our method is to conduct a semi-supervised learning in the selftraining framework by gradually labeling unlabeled data and repeatedly re-training using the most confidently classified instance pairs. An advantage of our method is that it is able to mine and utilize the data inform...

In the paper Pedersen, Nielsen, and Andersen [5] we developed an algorithm for ranking n×n assignments using reoptimization and compare our algorithm with other algorithms with the same time complexity. However, as pointed out by Dr. A. Volgenant, we unfortunately missed one available implementation written by Miller, Stone, and Cox [3] in IEEE Transactions on Aerospace and Electronic Systems. ...

Termination of loops can be inferred from the existence of linear ranking functions. We already know that the existence of these functions is PTIME decidable for simple rational loops. Since very recently, we know that the problem is coNP-complete for simple integer loops. We continue along this path by investigating programs dealing with floating-point computations. First, we show that the pro...

A node k-ranking of a graph G = (V, E) is a proper node coloring C: V  {1, 2, ..., k} such that any x-y path in G with C(x) = C(y) contains an internal node z with C(z) C(x). In the on-line version of this problem, the nodes v1, v2,..., vn are coming one by one in an arbitrary order; and only the edges of the induced subgraph G[{v1, v2,..., vi}] are known when the color of vi has to be chosen...

A ranking on a graph is an assignment of positive integers to its vertices such that any path between two vertices with the same label contains a vertex with a larger label. The rank number of a graph is the fewest number of labels that can be used in a ranking. The rank number of a graph is known for many families, including the ladder graph P2 × Pn. We consider how ”bending” a ladder affects ...

One can use the Leiden Rankings for grouping research universities by considering universities which are not significantly different as a homogeneous set. Such groupings reduce the complexity of the rankings without losing information. We pursue this classification using both statistical significance and effect sizes of differences among 902 universities in 54 countries; we focus on the UK, Ger...

The exchangeability assumption as defined in this paper on ranking functions seems intuitively natural, and indeed, specific ranking functions previously proposed in the literature are all exchangeable. While pointwise ranking functions are vacuously exchangeable, we now discuss two specifically listwise ranking functions previously proposed by [22] and [26] in light of our representation theor...

This paper is a continuation of the study of surprise as a base for constructing qualitative calculi for representing and reasoning about uncertain knowledge. Here, we further elaborate on κ, a qualitative ranking function which we developed in (Ibrahim, Tawfik, and Ngom 2009b) and which constructs qualitative ranks for events by obtaining the order of magnitude abstraction of the degree of sur...

In this paper, we provide an overview of the NTCIR IMine task, which is a core task of NTCIR-11 and also a succeeding work of INTENT@NTCIR-9 and INTENT2@NTCIR-10 tasks. IMine is composed of a subtopic mining (SM) task, a document ranking (DR) task and a TaskMine (TM) pilot task. 21 groups from Canada, China, Germany, France, Japan, Korea, Spain, UK and United States registered to the task, whic...