Auto Social Poster automatically submits each WordPress blog post to 34 social bookmarking sites.

Heart Hand Made Blog Feature

I was watching a rerun of Letterman today with a spoof called “Vatican Sin Update”. I thought it was just one of their antics but something also told me it could also be true, so off to Google I went.I was Catholic for 20 years and I live in a small town where people knows who’s not going to confession or abandoned the flock. The Catholic church is too influential here that people who are not Catholic can only have a chance to win an election if they are rich or backed up by the rich that’s why I still have days when I have the urge to know what is happening around in Vatican.

The paper describes a framework where video and image transcoding is driven by semantic information about the scenes being coded. Scenes are classified into highlight classes by means of finite state machines.These classes correspond to the concepts of a given ontology (e.g. soccer ontology or domotic security ontology) and are associated to a set of events that activate the detection process. Depending on the user preferences each highlight class is assigned a weight that drives image re-coding, with more quality for higher weights. As a result a video or image stream can be selectively and adaptively compressed depending on the kind of activity detected inside the stream. the authors introduce 2 examples on which they perform evaluations. The first about soccer and the second about indoor surveillance for domotic environments. Scene detection through FSM proved to be very good reaching a recognition rate always over 90%.

Dynamic adaptation has been an essential requirement for more and more business systems. Some research works have focused on the structural or behavioral changes of adaptive programs. There are also some works on adaptive components, with the emphasis on separation between control flow and basic functions of components. In these works, a business model for the domain is always missing, so a comprehensible business view of adaptations is unavailable for the user. In this paper, we propose a feature-oriented adaptive component model, which introduces the ontology-based feature model proposed in our previous work on feature-based domain modeling to provide both the business view for the user and adaptation basis for the system. Furthermore, the ontology-based model provides unambiguous terminology for both the business view and the component specification, which ensures the consistency between them. This feature-oriented adaptive component model has another characteristic of the micro control flow within the component, which enables the adaptation of the component behavior, including interaction sequence and style. The adaptive component model has been applied in our intelligent connector based framework for dynamic architecture, so a case study on the adaptive version of JPS (Java Pet Store) is illustrated to show the advantages of the component model

Lexical semantic classes of verbs play an important role in structuring complex predicate information in a lexicon, thereby avoiding redundancy and enabling generalizations across semantically similar verbs with respect to their usage. Such classes, however, require many person-years of expert effort to create manually, and methods are needed for automatically assigning verbs to appropriate classes. In this work, we develop and evaluate a feature space to support the automatic assignment of verbs into a well-known lexical semantic classification that is frequently used in natural language processing. The feature space is general – applicable to any class distinctions within the target classification; broad – tapping into a variety of semantic features of the classes; and inexpensive – requiring no more than a POS tagger and chunker. We perform experiments using support vector machines (SVMs) with the proposed feature space, demonstrating a reduction in error rate ranging from 48% to 88% over a chance baseline accuracy, across classification tasks of varying difficulty. In particular, we attain performance comparable to or better than that of feature sets manually selected for the particular tasks. Our results show that the approach is generally applicable, and reduces the need for resource-intensive linguistic analysis for each new classification task. We also perform a wide range of experiments to determine the most informative features in the feature space, finding that simple, easily extractable features suffice for good verb classification performance.