This paper concerns a method of selecting a subset of features for a sequential logit model. Tanaka and Nakagawa (2014) proposed a mixed integer quadratic optimization formulation for solving the problem based on a quadratic approximation of the logistic loss function. However, since there is a significant gap between the logistic loss function and its quadratic approximation, their formulation may fail to find a good subset of features. To overcome this drawback, we apply a piecewise-linear approximation to the logistic loss function. Accordingly, we frame the feature subset selection problem of minimizing an information criterion as a mixed integer linear optimization problem. The computational results demonstrate that our piecewise-linear approximation approach found a better subset of features than the quadratic approximation approach.
Carnes and Shmoys  presented a 2-approximation algorithm for the minimum knapsack problem. We extend their algorithm to the minimum knapsack problem with a forcing graph (MKPFG), which has a forcing constraint for each edge in the graph. The forcing constraint means that at least one item (vertex) of the edge must be packed in the knapsack. The problem is strongly NP-hard, since it includes the vertex cover problem as a special case. Generalizing the proposed algorithm, we also present an approximation algorithm for the covering integer program with 0-1 variables.
Modularity proposed by Newman and Girvan is the most commonly used measure when the nodes of a graph are grouped into communities consisting of tightly connected nodes. We formulate the modularity maximization problem as a set partitioning problem, and propose an algorithm for the problem based on the linear programming relaxation. We solve the dual of the linear programming relaxation by using a cutting plane method. To mediate the slow convergence that cutting plane methods usually suffer, we propose a method for finding and simultaneously adding multiple cutting planes.