Predicting Apps usage has become an important task due to the proliferation of Apps, and complex Apps. However, previous research works utilized a considerable number different sensors as training data infer usage. To save energy consumption for predicting usages, only temporal information is considered in this paper. We propose Temporal-based Predictor (abbreviated TAP) dynamically predict which are most likely be used. First, we extract three features, global feature, periodical feature...

Location prediction has attracted a significant amount of research effort. Given an object's recent movements and future time, the goal location is to predict this object at time specified. Prior works have elaborated on mining association relationships among regions, in which objects frequently appear, locations. Association regions are represented as rules. By exploring prior able good accuracy for prediction. However, with large trajectory data produced, huge rules expected. Furthermore,...

Due to the proliferation of mobile applications (abbreviated as Apps) on smart phones, users can install many Apps facilitate their life. Usually, browse by swiping touch screen and are likely spend much time browsing Apps. In this paper, we design an AppNow widget that is able predict users' usage. Therefore, could simply execute from widget. The main theme paper construct temporal profiles which identify relation between usage times. light Apps, predicts a list most be used at current...

Recently, with the advent of location-based social networking services (LBSNs), travel planning and location-aware information recommendation based on LBSNs have attracted much research attention. In this paper, we study impact relations hidden in LBSNs, i.e., The influence friends. We propose a new influence-based user recommender framework (SIR) to discover potential value from reliable users (i.e., Close friends experts). Explicitly, our SIR is able infer influential an LBSN. claim...

The recent build-up network of Automatic Identification System (AIS) equipped on vessels provides a rich source vessel movement information. AIS is originally designed for automatically exchanging navigation information, such as their unique identification, position, course, and speed, with nearby terrestrial receivers to affect collision avoidance safety control. collected sequences logs can be considered maritime trajectory data, i.e., the location points timestamps. This vast amount data...

By the rise of mobile devices, trajectory data could be easily collected and used in several applications, like destination prediction, public transportation optimization, travel route recommendation. However, due to spatio-temporal nature, raw usually contain redundant movement information. This observation motivates simplication approaches which discard some points with preserving specific features, such as position direction so on. Most existing simplifications ignore importance velocity...

Thanks for the common use of Automatic Identification System (AIS) network has made a large number maritime traffic data to be available. Ships equipped with AIS automatically exchange navigational information nearby ships and terrestrial receivers facilitate tracking monitoring ships' location movement collision avoidance control. Obviously, increasing amount shipping traffic, collisions are one growing safety concerns in situation awareness. To understand situations can help managers...

Collision-free is one of the major safety concerns for maritime traffic management. To analyze collision data and understand cause can contribute improvement However, real collisions not always available to analyze. Based on a massive AIS trajectory collected, we focus mining ships' movement behaviors those may bring possible if they do take any avoidance, called Maritime Traffic Conflict. Even though conflict non-accident incident, have similar navigational analysis. Thus, propose...

The improvement of collision avoidance for ship navigation in encounter situation is an important topic maritime traffic safety. Most research on has focused planning a safe path to keep away from the approaching under requirements International Regulations Preventing Collision at Sea (COLREGs). However, specific anti-collision actions are actually carried out by navigators' own experience according local situation.

With the popularity of location-based social networks (LBSNs), users would like to share their check-ins with friends for more interactions. These check-in records reflect not only when and where they are, but also what are doing. If we can capture relations location, time, activity factors in LBSNs, platforms provide personalized services users. In this paper, aim infer individual mobility based on LBSNs. For these two inference problems, analyze records, utilize Bayesian network represent...

Due to the proliferation of mobile applications(abbreviated as Apps) on smart phones, users can install many Apps facilitate their life. Usually, browse Appsby swiping touch screen and are likely spend much time browsing Apps. In this paper, we design an AppNow widget that is able predict users' usage. Therefore, could simply execute from widget. The main theme paper construct temporal profiles which identify relation between usage times. light Apps, predicts a list most be used at current...

The avoidance of collisions between ships in encounter situations is crucial to maritime traffic safety. Most research on collision has focused planning a safe path by which avoid approaching accordance with the requirements laid out International Regulations for Preventing Collision at Sea (COLREGs). resulting solution provides reference navigator movements collisions. Nonetheless, specific anti-collision actions are generally based experience navigator. This study differed from existing...

As security requirements in coastal water and sea ports, maritime surveillance increases the duty. In this research, we focus on trajectory data to explore movement behavior for anomaly detection traffic. Trajectory records moving objects' true provides opportunity discover detection. The multidimensional outlying features are first identified defined. To deal with uncertain property of trajectory, a modeling is developed from historical trajectories build model Then, our ongoing work...

With the rapid increase in global maritime shipping, there is a great demand for technology of traffic monitoring to detect inappropriate encountering interaction between ships and prevent ship collision accidents. The Automatic Identification System (AIS) network makes it possible collect large volume data investigate avoidance behavior real-world ships. Most systems are based on expert simulations International Regulations Preventing Collisions at Sea (COLREGs). Those regulations outline...

Existing fake news research relies on propagation or metadata. Waiting for structure to be enough is a waste of time. Hoping reliable metadata information also because all data can forged. The most natural way human when verifying through the content itself. In social media, circulating are in minimal which consist image and its text caption. We propose FakeCLIP examine whether caption truly describes corresponding not. As far as we know, first one tackle using approach. found mixed...

In this paper, we proposed a hybrid recommendation model to tackle two challenges in the system. First, many of products have been browsed frequently but may not consequentially be ordered. As result, users' actions directly considered as preference on specific item. Second, popularity sold has highly skewed distribution which results cold start problem recommendation. order extract knowledge from implicit feedback, develop neighborhood structure users and behaviors multi-behavior...