A Hybrid Data Analytics Framework with Sentiment Convergence and Multi-Feature Fusion for Stock Trend Prediction

Stock market analysis plays an indispensable role in gaining knowledge about the stock market, developing trading strategies, and determining the intrinsic value of stocks. Nevertheless, predicting stock trends remains extremely difficult due to a variety of influencing factors, volatile market news, and sentiments. In this study, we present a hybrid data analytics framework that integrates convolutional neural networks and bidirectional long short-term memory (CNN-BiLSTM) to evaluate the impact of convergence of news events and sentiment trends with quantitative financial data on predicting stock trends. We evaluated the proposed framework using two case studies from the real estate and communications sectors based on data collected from the Dubai Financial Market (DFM) between 1 January 2020 and 1 December 2021. The results show that combining news events and sentiment trends with quantitative financial data improves the accuracy of predicting stock trends. Compared to benchmarked machine learning models, CNN-BiLSTM offers an improvement of 11.6% in real estate and 25.6% in communications when news events and sentiment trends are combined. This study provides several theoretical and practical implications for further research on contextual factors that influence the prediction and analysis of stock trends.

Design of a Food Recommendation System using ADNet algorithm on a Hybrid Data Mining Process

Hybrid data mining processes are employed in recent days on several applications to achieve a better prediction and classification rate along with customer satisfaction. Hybrid data mining processes are the combination of different form of data considered for a neural network decision. In some cases, the different form of data represents image along with numerical data. In the proposed work, a food recommendation system is developed with respect to the flavour taste of the customer and considering the review comments of previous customers. The suggestions given by the users are taken into account as a feedback layer in the neural network for fine tuning the accuracy of the prediction process. The architectural design of the proposed model is employed with an ADNet (Adaptively Dense Convolutional Neural Network) algorithm to enable the usage of low range features in an efficient way. To verify the performance of the developed model, a pizza flavour recommender dataset is employed in the work for analysis. The experimental work analysis indicates that the ADNet algorithm works in a better way on a hybrid data analysis than the traditional DenseNet and ResNet algorithms.

Determination of Student Subjects in Higher Education Using Hybrid Data Mining Method with the K-Means Algorithm and FP Growth

The large number of courses offered in an educational institution raises new problems related to the selection of specialization courses. Students experience difficulties and confusion in determining the course to be taken when compiling the study plan card. The purpose of this study was to cluster student value data. Then the values that have been grouped are seen in the pattern (pattern) of the appearance of the data based on the values they got previously so that students can later use the results of the patterning as a guideline for taking what skill courses in the next semester. The method used in this research is the K-Means and FP-Growth methods. The results of this rule can provide input to students or academic supervisors when compiling student study plan cards. Lecturers and students can analyze the right specialization subject by following the pattern given. This study produces a pattern that shows that the specialization course with the theme of business information systems is more followed by students than the other 2 themes

Case Studies of Digitalized Locate the Remaining Oil Workflows Powered by Hybrid Data & Physics Methods

Mature fields already account for about 70% of the hydrocarbon liquids produced globally. Since the average recovery factor for oil fields is 30 to 35%, there is substantial quantities of remaining oil at stake. Conventional simulation-based development planning approaches are well established, but their implementation on large, complex mature oil fields remains challenging given their resource, time, and cost intensity. In addition, increased attention towards reduce carbon emissions makes the case for alternative, computationally-light techniques, as part of a global digitalisation drive, leveraging modern analytics and machine learning methods. This work describes a modern digital workflow to identify and quantify by-passed oil targets. The workflow leverages an innovative hybrid physics-guided data-driven, which generates historical phase saturation maps, forecasts future fluid movements and locate infill opportunities. As deliverables, a fully probabilistic production forecast is obtained for each drilling location, as a function of the well type, its geometry, and position in the field. The new workflow can unlock remaining potential of mature fields in a shorter time-frame and generally very cost-effectively compared to the advanced dynamic reservoir modelling and history-match workflows. Over the last 5 years, this workflow has been applied to more than 30 mature oil fields in Europe, Africa, the Middle East, Asia, Australia, and New Zealand. Three case studies’ examples and application environments of applied digital workflow are described in this paper. This study demonstrates that it is now possible to deliver digitalized locating the remaining oil projects, capturing the full uncertainty ranges, including leveraging complex multi-vintage spatial 4D datasets, providing reliable non-simulation physics-compliant data-driven production forecasts within weeks.