Accurately classifying colorectal polyps, or differentiating malignant from benign ones, has a significant clinical impact on early detection and identifying optimal treatment of colorectal cancer. Convolution neural network (CNN) has shown great potential in recognizing different objects (e.g. human faces) from multiple slice (or color) images, a task similar to the polyp differentiation, given a large learning database. This study explores the potential of CNN learning from multiple slice (or feature) images to differentiate malignant from benign polyps from a relatively small database with pathological ground truth, including 32 malignant and 31 benign polyps represented by volumetric computed tomographic (CT) images. The feature image in this investigation is the gray-level co-occurrence matrix (GLCM). For each volumetric polyp, there are 13 GLCMs, computed from each of the 13 directions through the polyp volume. For comparison purpose, the CNN learning is also applied to the multi-slice CT images of the volumetric polyps. The comparison study is further extended to include Random Forest (RF) classification of the Haralick texture features (derived from the GLCMs). From the relatively small database, this study achieved scores of 0.91/0.93 (two-fold/leave-one-out evaluations) AUC (area under curve of the receiver operating characteristics) by using the CNN on the GLCMs, while the RF reached 0.84/0.86 AUC on the Haralick features and the CNN rendered 0.79/0.80 AUC on the multiple-slice CT images. The presented CNN learning from the GLCMs can relieve the challenge associated with relatively small database, improve the classification performance over the CNN on the raw CT images and the RF on the Haralick features, and have the potential to perform the clinical task of differentiating malignant from benign polyps with pathological ground truth.

We propose Topic Anchoring-based Review Summarization (TARS), a two-step extractive summarization method, which creates review summaries from the sentences that represent the most important aspects of a review. In the first step, the proposed method utilizes Topic Aspect Sentiment Model (TASM), a novel sentiment-topic model, to identify aspects of sentiment-specific topics in a collection of reviews. The output of TASM is utilized in the second step of TARS to rank review sentences based on how representative of the most important review aspects their words are. Qualitative and quantitative evaluation of review summaries using two collections indicate the effectiveness of structuring review summaries around aspects of sentiment-specific topics.

We study the problem of minimizing total completion time in 2-stage flowshop with availability constraint.This problem is NP-hard in the strong sense even if both machines are always available.With availability constraint, although a bulk of research papers have studied the makespan minimization problem, there is no research done on the total completion time minimization.This paper is the first attempt to tackle this problem.We focus on the case that there is a single unavailable interval on the first machine only.We show that several special cases can be solved optimally or approximated within a constant factor.For the general case, we develop some lower bounds and dominance rules.Then we design and implement a branch and bound algorithm.We investigate the effectiveness of different lower bounds and the dominance rules by computational experiments.We also study how the start time and the duration of the unavailable interval affects the efficiency of the branch and bound algorithm.

We consider the problem of scheduling a set of n unit-execution-time (UET) tasks, with precedence constraints, on m ≥ 1 parallel and identical processors so as to minimize the mean flow time. For two processors, the Coffman--Graham algorithm gives a schedule that simultaneously minimizes the mean flow time and the makespan. The problem becomes strongly NP-hard for an arbitrary number of processors, although the complexity is not known for each fixed m ≥ 3. For arbitrary precedence constraints, we show that the Coffman--Graham algorithm gives a schedule with a worst-case bound no more than 2, and we give examples showing that the bound is tight. For intrees, the problem can be solved in polynomial time for each fixed m ≥ 1, although the complexity is not known for an arbitrary number of processors. We show that Hu's algorithm (which is optimal for the makespan objective) yields a schedule with a worst-case bound no more than 1.5, and we give examples showing that the ratio can approach 1.308999.

A fundamental problem in scheduling theory is that of scheduling a set of n tasks, with precedence constraints, on $m \ge 1$ identical and parallel processors so as to minimize the makespan (schedule length). In the past, research has focused on the setting whereby all tasks are available for processing at the beginning (i.e., at time t = 0). In this article we consider the situation where tasks, along with their precedence constraints, are released at different times, and the scheduler has to make scheduling decisions without knowledge of future releases. In other words, the scheduler has to schedule tasks in an online fashion. We consider both preemptive and nonpreemptive schedules. We show that optimal online algorithms exist for some cases, while for others it is impossible to have one. Our results give a sharp boundary delineating the possible and the impossible cases.