This paper presents a comparative analysis of various methods for automatic sentiment labeling. We investigate the performance of three semi-supervised learning methods alongside a state-of-the-art zero-shot learning model on four datasets. Our experiments reveal the strengths and weaknesses of each approach, particularly in scenarios with limited labeled data. While traditional semi-supervised methods demonstrate exemplary performance in leveraging unlabeled data, zero-shot classification offers a compelling alternative without requiring labeled instances for the target task. However, our findings also highlight significant limitations of zero-shot learning, mainly when applied to datasets with complex textual structures. These insights underscore the need for careful method selection based on the specific characteristics and constraints of the sentiment analysis task.

This paper reports on a study conducted to incorporate code quality into an introductory Java programming course. The discussion focuses on the strategies we used to teach coding standards necessary for writing high-quality code. We also present data and an analysis, investigating code quality issues identified through CheckStyle and PMD in students' code submissions. Our analysis revealed that 42% of code quality issues are related to code formatting and documentation, with an average of 363.19 issues per Thousand Lines of Code ( KLOC ) and 49.96 issues per KLOC , respectively. The analysis also revealed the presence of error-prone and other best practices issues. This analysis provides insight into the effectiveness of these teaching strategies.

This paper introduces the development of a mobile application employing Natural Language Processing (NLP) techniques to provide content-based recommendations for Sci-Fi books. The mobile application integrates two distinct NLP techniques: Doc2Vec for rapid keyword searches and RoBERTa to enhance the understanding of book themes and ideas. This combination enables the recommender system to offer personalized book recommendations tailored to individual user interests, enhancing the reading experience.

This study investigates the potential use of Natural Language Processing (NLP) techniques to analyze coding standards violations within the context of an introductory programming course. In particular, the study evaluates the effectiveness of various advanced text embedding techniques, including Bag of Words (BOW), Doc2Vec, and BERT, in clustering coding standards violations. This study aims to determine which text embedding techniques contribute to the most accurate clustering of errors. Our findings highlight the superiority of Doc2Vec embeddings in effectively clustering related errors compared to the alternative techniques.

In order to address current water challenges, scientific research on water-related issues is crucial. However, traditional techniques for selecting research topics, such as literature reviews and expert opinions, can be time-consuming and may not provide a comprehensive overview of available information. We propose using Natural Language Processing (NLP) techniques in this study to extract, align, and compare water research topics from different corpora. We apply these techniques to the research paper abstracts from the New Mexico Water Resources Research Institute (NMWRRI) and the U.S. Geological Survey (USGS) to assess these institutions’ current research interests and identify potential new research directions. We use a Latent Dirichlet Allocation (LDA) model for topic extraction and a Word2Vec model for topic alignment. This study highlights the benefits of using NLP techniques to analyze trends and identify novel research directions in water studies.

The ability to accurately estimate crop numbers for various varieties of vegetables makes crop classification an essential component of agriculture. In order to meet the anticipated demand in the future, dealers can predict the number of crops sold on the market. For crop classification, scientists have recently used a variety of data sources, including optical (Polarimetric) and radar sensing imaging. They have, however, run into difficulties when working with huge, high-dimensional, and imbalanced datasets. Some researchers have used a 3-stacked generalization strategy to address these issues. This approach was difficult to train and tackling massive data and imbalanced data concerns head-on. They were able to get an F1-score of 85% with this technique. We used the regression feature selection technique and data sampling in our work, using only 20% of the total data as the training dataset. The effectiveness of our classification methods was significantly improved and the training time was much decreased by these preprocessing techniques. We specifically acquired a remarkable F1-score of 99% after just 42 seconds of training utilizing the Random Forest algorithm. We also achieved a respectable F1-score of 97% in under 7 seconds using Linear SVC. We also demonstrate that good performance may be achieved by using one day's worth of radar data with only 38 features to cut costs and time. The results show that we can use only 16 features to get high performance. In our research, we found that employing radar data produces outcomes that are higher performing and more accurate than using optical data. We have out extensive experiments to show the potency of our methods.

Local Memory Store (LMStore) is a novel scratchpad memory (SPM) design, with recent research evaluation showing its capability for improving program performance. However, the performance of LMStore depends on its memory layout decided by its allocation scheme. In this paper, we evaluate the impact of SPM allocation on LMStore performance. Our experimental results, using benchmarks from the Malardalen WCET benchmark suite executing on LMStore architecture modeled in the PyCacheSim simulator, demonstrate that LMStore with a stack distance-based SPM allocation scheme significantly improves data movement by an average of 44.46% compared to a Cache-only architecture, and by an average of 23.89% compared to LMStore with a frequency-based SPM allocation scheme.

Local Memory Store (LMStore) is a novel hardware-controlled, compiler-managed Scratchpad memory (SPM) design [1], with an initial research evaluation that showed its possibility for improving program performance. This initial evaluation was performed over memory traces prior to the development of compiler support for LMStore. In this paper, we present compiler support for the LMStore design, and present experimental results that better evaluate LMStore performance. Experimental results on benchmarks from Malardalen benchmark suite [2] executing on the LMStore architecture modeled in Multi2Sim demonstrate that a hybrid LMStore-Cache architecture improves execution time by an average of 19.8 %, compared to a conventional cache-only architecture.

Distance learning has dramatically increased in recent years because of advanced technology. In addition, numerous universities had to offer courses in online mode in 2020 and 2021 because of the COVID-19 pandemic. However, there are more challenges in distance learning than in the traditional learning method (e.g., feedback and interaction). Recently, researchers started using simple EEG headsets to identify confused students during online courses based on machine learning approaches. However, they faced unpleasant accuracy using traditional machine learning algorithms or non-deep neural networks. In this paper, we present a data-driven approach based on a multi-view deep learning technique called CSDLEEG to identify confused students. We employ the students' demographic information and EEG signals to feed our novel neural networks. The results show that our proposed approach is superior to state-of-the-art methods for 98% accuracy and 98% F1-score.