In this paper we present two deep-learning systems that competed at SemEval-2017 Task 4 “Sentiment Analysis in Twitter”. We participated all subtasks for English tweets, involving message-level and topic-based sentiment polarity classification quantification. use Long Short-Term Memory (LSTM) networks augmented with kinds of attention mechanisms, on top word embeddings pre-trained a big collection Twitter messages. Also, text processing tool suitable social network messages, which performs...

Recent efforts in spatial and temporal data models database systems have attempted to achieve an appropriate kind of interaction between the two areas. This paper reviews different types spatio-temporal that been proposed literature as well new theories concepts emerged. It provides overview previous achievements within domain critically evaluates various approaches through use a case study construction comparison framework. comparative review is followed by comprehensive description lines...

Trajectory database (TD) management is a relatively new topic of research, which has emerged due to the explosion mobile devices and positioning technologies. similarity search forms an important class queries in TD with applications trajectory data analysis spatiotemporal knowledge discovery. In contrast related works make use generic metrics that virtually ignore temporal dimension, this paper we introduce framework consisting set distance operators based on primitive (space time) as well...

Moving Object Databases (MOD), although ubiquitous, still call for methods that will be able to understand, search, analyze, and browse their spatiotemporal content. In this paper, we propose a method trajectory segmentation sampling based on the representativeness of (sub)trajectories in MOD. order find most representative subtrajectories, following methodology is proposed. First, novel global voting algorithm performed, local density similarity information. This applied each segment...

Indoor Positioning Systems (IPS) have recently received considerable attention, mainly because GPS is unavailable in indoor spaces and consumes energy. On the other hand, predominant Smartphone OS localization subsystems currently rely on server-side processes, allowing service provider to know location of a user at all times. In this paper, we propose an innovative algorithm for protecting users from tracking by service, without hindering provisioning fine-grained updates continuous basis....

Data stream clustering is a hot research area due to the abundance of data streams collected nowadays and need for understanding acting upon such sort data. Unsupervised learning (clustering) comprises one most popular mining tasks gaining insights into Clustering challenging task, while over involves additional challenges as single pass constraint raw fast response. Moreover, dealing with an infinite changing implies that model extracted also subject evolution time. Several surveys exist...

Forecasting vessel locations is of major importance in the maritime domain, with applications safety, logistics, etc. Nowadays, tracking has become possible largely due to increased GPS-based data availability. This paper introduces a novel Vessel Location (VLF) framework, based on Long-Short Term Memory (LSTM) Neural Networks, aiming perform effective location forecasting time horizons up 60 minutes, even for vessels not recorded past. The proposed VLF framework specially designed handling...

In this paper, we propose a novel scheme for efficient content-based medical image retrieval, formalized according to the PAtterns Next generation DAtabase systems (PANDA) framework pattern representation and management. The proposed involves block-based low-level feature extraction from images followed by clustering of space form higher-level, semantically meaningful patterns. is realized an expectation-maximization algorithm that uses iterative approach automatically determine number...

The flow of data generated from low-cost modern sensing technologies and wireless telecommunication devices enables novel research fields related to the management this new kind implementation appropriate analytics for knowledge extraction. In work, we investigate how traditional cube model is adapted trajectory warehouses in order transform raw location into valuable information. particular, focus our on three issues that are critical warehousing: (a) reconstruction procedure takes place...

Mining trajectory databases (TD) has gained great interest due to the popularity of tracking devices. On other hand, inherent presence uncertainty in TD (e.g., GPS errors) not been taken yet into account during mining process. In this paper, we study effect clustering and introduce a three-step approach deal with it. First, propose an intuitionistic point vector representation trajectories that encompasses underlying effective distance metric cope uncertainty. Second, devise CenTra, novel...

Accurate prediction of Public Transport (PT) mobility is important for intelligent transportation. Nowadays, data have become increasingly available with the General Transit Feed Specification (GTFS) being format PT agencies to disseminate such data. Estimated Time Arrival (ETA) crucial public, as well agency logistics, route-optimization, maintenance, etc. However, PT-ETA a challenging task, due complex and non-stationary urban traffic. This work introduces novel data-driven approach...

We present HERMES, a prototype system based on powerful query language for trajectory databases, which enables the support of aggregative Location-Based Services (LBS). The key observation that motivates HERMES is more knowledge in hand about mobile user, better exploitation advances spatio-temporal processing providing intelligent LBS. fully incorporated into state-of-the-art Object-Relational DBMS, and its demonstration illustrates flexibility usefulness delivering custom-defined

Challenged by real-world clustering problems this paper proposes a novel fuzzy scheme of datasets produced in the context intuitionistic set theory. More specifically, we introduce variant Fuzzy C-Means (FCM) algorithm that copes with uncertainty and similarity measure between sets, which is appropriately integrated algorithm. We describe an fuzzification colour digital images upon applied proposed scheme. The experimental evaluation shows it can be more efficient effective than...