Although deep learning (DL) algorithms have been proved to be effective in diverse research domains, their application in developing models for tabular data remains limited. Models trained on tabular data demonstrate higher efficacy using traditional machine learning models than DL models, which are largely attributed to the size and structure of tabular datasets and the specific application contexts in which they are utilized. Thus, the primary objective of this paper is to propose a method to use the supremacy of Stacked Bidirectional LSTM (Long Short-Term Memory) deep learning algorithms in pattern discovery incorporating tabular data with customized 3D tensor modeling in feeding neural networks. Our findings are empirically validated using six diverse, publicly available datasets each varying in size and learning objectives. This paper proves that the proposed model based on time-sequence DL algorithms, which were generally described as inadequate when dealing with tabular data, yields satisfactory results and competes effectively with other algorithms specifically designed for tabular data. An additional benefit of this approach is its ability to preserve simplicity while ensuring fast model training also with large datasets. Even with extremely small datasets, models can be applied to achieve exceptional predictive results and fully utilize their capacity.

Clustering users on social media based on text involves grouping individuals with similar text patterns, language usage, or content interests. This text-based clustering provides insights into user preferences, enables personalized content recommendations, and facilitates understanding of social networking trends and user engagement. However, traditional text clustering methods rely heavily on language-specific features. This limits their applicability in multilingual media environments where linguistic diversity prevails. In this paper, the problem of clustering users on social networks, specifically focusing on text-based clustering independent of the language in which the text is written, is addressed. A practical methodology is presented, outlining an iterative procedure for clustering based solely on language-independent features such as emojis, hashtags, URLs, text length, and punctuation count. The effectiveness of the language-independent clustering approach is compared with the usual text based clustering approach. Comparison of these results shows that for the used dataset, the proposed clustering method using language independent features gives higher quality results than text clustering.

ABSTRACT Transportation management, as a part of the supply chain management, is a complex process that consists of planning and delivering goods to customers. The paper presents a complete multi-phase intelligent and adaptive transportation management system, which includes data collection, parameter tuning, and the heuristic algorithm based on the Tabu search for vehicle routing. The paper describes the procedure for collecting Global Positioning System (GPS) data and analyzing the compliance with the proposed routes based on the data collected. The described routing algorithm is powerful and supports many real-world limitations. An algorithm for the anomaly detection in the GPS data is presented as well as the usage of collected GPS data to improve the future results of the algorithm. The concept was implemented and tested on real data in some of the largest distribution companies in Bosnia and Herzegovina. The proposed approach resulted with more than satisfactory results in real-world application.

The past decade was marked, among other things, by the rapid growth of social networks. These networks collect personal data about their users - their photographs, interests, friends, locations, website visits, clicks, status updates and much more. A large number of users and a big collection of various data collected about the users make social media networks an abundant source of data that can be analyzed and used for targeted marketing, social phenomena analysis, generating different statistics and so on. In this paper we will use the potential of the tool RapidMiner in order to collect data from the social media network Twitter using the AYLIEN extension, preparing the data and applying sentiment analysis, which will give insight into the general atmosphere surrounding the actions of the current USA president Donald Trump

Traffic is a complex system, in terms of functioning as well as organization. Traffic stop, i.e. failure to perform the required transport function from point A into point B, can be caused by variety of circumstances, and one of them is the occurrence of traffic accident. Therefore, it is of a great importance to pay attention to increasing the traffic safety level throughout the implementation of various measures. Traffic safety is one of the most important links of the traffic system. This paper analyzes in details the impact of the road and its environment, vehicles and drivers on traffic accidents. The model of classification of traffic accidents causes has been implemented on the basis of similarities of drivers, vehicles and road characteristics. k-means clustering algorithm has been used for this purpose. According to data available, time-series prediction model has been implemented in the second part of the work, for prediction of traffic accidents in analyzes regions. Results in both implemented cases are more than satisfactory.