An Ensemble Approach for Emotion Cause Detection with Event Extraction and Multi-Kernel SVMs

In this paper, we present a new challenging task for emotion analysis, namely emotion cause extraction.In this task, we focus on the detection of emotion cause a.k.a the reason or the stimulant of an emotion, rather than the regular emotion classification or emotion component extraction. Since there is no open dataset for this task available, we first designed and annotated an emotion cause dataset which follows the scheme of W3 C Emotion Markup Language. We then present an emotion cause detection method by using event extraction framework,where a tree structure-based representation method is used to represent the events. Since the distribution of events is imbalanced in the training data, we propose an under-sampling-based bagging algorithm to solve this problem. Even with a limited training set, the proposed approach may still extract sufficient features for analysis by a bagging of multi-kernel based SVMs method. Evaluations show that our approach achieves an F-measure 7.04%higher than the state-of-the-art methods.

Climatic change: Causal correlations over the last 240 Ma

The climate of the Earth has been oscillating between mega warm periods and mega cold periods for 3,000 Ma. Each mega cold period included alternating major warm and cold events. The present mega cold period commenced about 44 Ma in the polar re- gions as the seas cooled following the loss of the circum-equatorial ocean. Before then, a mega warm period lasted for more than 200 Ma. The frequency of the major cold events within the present mega cold period is increasing, with each continent being un- der the influence of a different set of climatic controls. There are many causes of these shifts in climate, ranging from fluctuating meridional ocean currents, rearrangement of tectonic plates, and changes in ocean gateways. These are enhanced by a combination of Milankovitch cycles and many other medium to small oscillations and cyclic controls that cause the daily, monthly, and season- al fluctuations in weather. Examples are given of how these can cause a change from cold to warm events, or vice versa, at pre- sent-day or mega scales, aided by eustatic changes in sea levels and changes in the distribution of air masses, sea ice, and snow.

Reliability evaluation of wireless sensor networks using an enhanced OBDD algorithm

An enhanced ordered binary decision diagram （EOBDD） algorithm is proposed to evaluate the reliability of wireless sensor networks （WSNs）, based on the considerations of the common cause failure （CCF） and a large number of nodes in WSNs. The EOBDD algorithm analyzes the common cause event （CCE） and the network structure when CCE takes place according to the stochastic graph and the CCF model of WSNs. After constructing the ordered binary decision diagram （OBDD） of the original network with node expansion, it uses a set of OBDD variables （SOV） to guide reliability computations along this OBDD. The two steps about OBDD can decrease the cost of OBDD constructions and storage. Furthermore, the efficient OBDD structure and Hash tables can greatly decrease redundant computations of isomorphs. The experiment results show that the EOBDD can be used to evaluate the reliability of WSN efficiently.