Deep Learning is a field included in to Artificial Intelligence. It allows computational models to learn multiple levels of abstraction with multiple processing layers. This Artificial Neural Networks gives state-of-art performance in various fields like Computer Vision, Speech recognition and different domain like bioinformatics. There are mainly three architectures of Deep Learning Convolution Neural Network, Deep Neural Network and Recurrent Neural Network which provides the higher level of representation of data at each next layer. Deep Learning is required to classify high dimensional data like images, audio, video and biological data.
Time Series Forecasting (TSF) is crucial in various real-world applications such as climate forecasting and electricity demand prediction. Unlike traditional datasets, time series data points are influenced by their past values, necessitating specialized techniques to model these sequential dependencies specifically addressing non-linear patterns, abrupt changes, and outliers. The latest advancements have significantly enhanced TSF using machine learning and other methods. However, forecasting extreme events remains challenging. Extreme values, although rare, have significant real-world impacts such as heavy rainfall, fluctuations in electricity demand, and traffic surges. This paper proposes a TXtreme framework that uses Long-Short memory network, feed-forward neural network, and transformer to improve time series forecasting under extreme values. The model also uses statistical methods to explain the distribution of time series values. Extensive experiments are conducted using datasets from different domains to show the robustness of the proposed methodology. Results, derived by testing TXtreme on five datasets of different domain, indicate that TXtreme significantly outperforms state-of-the-art methods in time series forecasting, with improvements of 5–25% in root means squared error or mean absolute error. The proposed framework enhances TSF capabilities and ensures better generalization ability in extreme event forecasting, potentially leading to improved decision-making in critical applications.
Objectives: To propose a method to handle large number of user and to improve the accuracy and quality of recommendation system. Methods/Statistical Analysis: This paper presents an effective method to identify user location based on his/her preference using Skyline query outline Dominated object. Dominance object suggests that an object falls under good or better in all dimension or good at least one dimension. Skyline query using Recommendation system has increased in recent years. Skyline query using recommendation system mainly used location-based services to find the nearest location, based on user preference. Location-based Services are information services and have a number of uses in social networking. Location-based Service finds the nearest location based on user preferences but not provide location based on similarity and rating. So, the user is not satisfied by the given result. Findings: To resolve above problem, the collaborative filtering technique, K-nearest neighbor algorithm and Ranking Scheme being used by us. Using Collaborative filtering technique, we find the similarity and rating of an item. Using K-nearest neighbor approach finds the nearest distance of the similar item and ranking technique being used by us, to choose the most nearest location. In this paper we take temporary dataset and mathematically evaluate our proposed system. Application/Improvements: In future, we will develop web tool which identify location and display result on map. We will also check user s' past movement history based on content based recommendation system. Skyline query using recommendation system is use various domain i.e. House Rent/buying, travel and tourism business.
1 January 2016 in International Journal Of Data Mining And Emerging Technologies
Skyline query and recommendation system is more trendy in multi-criteria decision-making system, data mining and database field. Skyline query is used by numerous applications in recommendation i.e customised, collaborative and knowledge-based recommendation. Using skyline query in recommendation, the user finds the closest location based on his/her preference. Skyline query outlines dominated object. The dominance object suggests that an object is good or better than all dimensions or it is good at least in one dimension. Skyline operator extends the traditional database system. Skyline query is lot of fashionable in geo-spatial area. This paper survey is more attention-grabbing and still evolving research area, in order that the reader will simply acquire an outline of state of-the-art. We have a tendency to review very well existing approaches that are applicable for geo-spatial location environment. Our analysis results in taxonomy of the existing approach. Finally, we tend to present fascinating research topic on skyline query using in recommendation that haven't explored.