Deploying Reasoning LLMS for Sentiment Analysis: Architecture Trade-Offs on Consumer Hardware

Sentiment analysis refers to the automated process of identifying the underlying emotional tone conveyed in a text. This task has become increasingly crucial today due to the exponential growth of opinion-oriented text generation on various online platforms including social media and e-commerce websites. The process of sentiment analysis has evolved through different phases like the use of handcrafted rules, sentiment-oriented dictionaries, machine learning, deep learning, transformer models, and large language models. These different phases have enriched the field of sentiment analysis and enabled it to get better insights into text for sentiment analysis. One of the most impactful developments in sentiment analysis was the creation of embedding’s for words or sentences, which are useful for representing unstructured data as structured data. These embedding are representing words or sentences as numerical vectors in a high-dimensional space and have proven effective in capturing the semantic relationships between words. By leveraging the power of embedding’s, sentiment analysis models can understand the complexities of human language and provide more precise insights into people's emotions and opinions. However, word embedding’s not inherently generated for sentiment analysis, which means that they are not inherently sentiment-oriented. Therefore, many researchers have focused on developing sentiment-oriented word embedding’s for sentiment words, but they have overlooked the importance of intensity words such as “little”, “high”, and “extremely”. These intensity words determine the level of sentiment expressed in text, which is particularly useful for fine-level sentiment analysis. To address this issue, a review proposes an intensity-aware word embedding development model and sentiment analysis with ML.

Natural language processing (NLP) is being widely used globally for a variety of value-creation tasks ranging from chat-bots and machine translations to sentiment and topic analysis and multilingual large language models (LLMs). However, most of the advances are initially implemented within the English language framework, and it takes time and resources to develop comparable resources in other languages. The advances in machine translations have enabled the rapid and effective conversion of content in global languages into English and vice-versa. This creates potential opportunities to apply English language NLP methods and tools to other languages via machine translations. However, although this idea is powerful, it needs to be validated and processes and best practices need to be developed and kept updated. The present research is an effort to contribute to the development of best practices and an evaluation framework. We present a systematic and repeatable state-of-the-art process to evaluate the viability of applying English language sentiment analysis tools to Italian text by using multiple English language machine translation mechanisms such that it can be easily extended to other languages.

Aim/Purpose: This paper focuses on app review analysis techniques, driven by the rapid advancement of the mobile app market and NLP techniques in optimizing mobile app user experiences. Background: Owing to technological advancements, app review analysis has rapidly evolved. This study examines both conventional and emerging techniques, including current advancements such as large language models (LLMs) in app review analysis. It provides an overview of the various methods used across different categories of app review analysis, comparing effective strategies for identifying user concerns and enhancing app functionality. Methodology: A systematic review was utilized based on two major standard guidelines, PRISMA and Kitchenham’s guidelines, for the period of 2014 to 2024. After defining the review protocol, papers were identified through keyword-based searches on six major online databases: Scopus, Web of Science, IEEE Xplore, ACM Digital Library, Science Direct, and Springer. Following screening and excluding papers based on defined quality criteria, 53 papers were considered for this study. The use of PRISMA ensures a transparent and reproducible review process, while Kitchenham’s guidelines provide a structured and rigorous approach for evaluating and synthesizing the literature. Contribution: This review study aims to evaluate the current state of knowledge on app review analysis techniques to improve mobile app user experiences. This study categorized the existing state-of-the-art papers into eight different categories, such as sentiment analysis, review classification, summarization, and prioritization, and examined challenges related to app review analysis. Furthermore, the study emphasizes the potential of LLMs for optimizing and automating app review analysis and provides future directions to address gaps in user-centric app development. Findings: Among the eight main categories defined in app review analysis, sentiment analysis is the most prevalent, followed by review classification and information extraction. Most studies use a combination of these categories to achieve a comprehensive goal. Prioritization techniques such as risk matrices, thumbs-up count-based approach, and anomaly detection are widely used to identify emerging issues. Extracting meaningful information and evaluating the proposed approach are the most common challenges identified. Novel LLMs, like Chat-GPT, significantly enhance review analysis by automating the process, improving feature extraction, and enabling context-aware review classification. Recommendations for Practitioners: The combination of conventional approaches and novel LLM-based methods can enhance both the efficiency and accuracy in identifying and addressing critical issues raised through mobile app user reviews. It effectively prioritizes user concerns by leveraging the strengths of both traditional preprocessing techniques and advanced LLMs. Recommendation for Researchers: Researchers are encouraged to explore the integration of emerging technologies like LLMs to enhance the of app review analysis, particularly in feature-specific sentiment analysis. Impact on Society: The results of this study contribute to enhancing the mobile app user experience through effective app review analysis, which improves user satisfaction and supports user-centered app development. This ultimately leads to a better mobile app ecosystem, benefiting both users and developers. Future Research: In the future, this research can be extended in multiple directions. Researchers can address the existing research gaps that LLMs have yet to address, particularly in prioritizing user concerns. Additionally, there is potential for further research on tool implementations focusing on identifying persistent issues through time series analysis by considering the app version and date of the app reviews. Moreover, there is a need to develop comprehensive frameworks that are more generalizable across different apps and categories, with a focus on identifying user concerns related to specific features.

In this cognitive era, vast amount of data are accumulated every day. Analysing such unstructured information and obtaining insights will be challenging. To address this, Large language models have been developed to support analysis of extensive data corpora. However, it tends to cause hallucination due to a lack of proper knowledge sources. If the analysis has to be performed with respect to the health care domain or finance domain, the challenge is raised because of the lack of domain specificity. COVID-19 sentiment analysis is one of the complex responsibilities of the government since it needs to know the opinions of people to take necessary measures. This paper presents COVID-19 retrieval augmented and fine-tuning (RAFT), a novel framework that includes the analysis of COVID-19 vaccine tweets through retrieval augmented-based approaches. This integrated domain-specific knowledge through a retrieval-augmented generation-based approach with external knowledge sources. We employed a transformer-based semantic approach in embedding generation via vector database. Furthermore, this framework exhibited generalizability when integrated with domain knowledge. It uses parameter efficient fine tuning with quantization to use a large language model with a reduced number of parameters, which will allow a model to be used in low-resource-constrained devices. This framework achieved an accuracy of 0.886 on the Twitter dataset containing tweets specific to Indian region and 0.912 on the Twitter dataset with tweets from Global region.

Sentiment analysis is a computational technique employed to extract and interpret subjective information from textual data. It involves the identification and classification of sentiments, opinions, and emotions expressed within the text. By analyzing linguistic cues, such as word choice, syntax, and sentiment lexicons, sentiment analysis can discern a range of emotions, from positive to negative, as well as more nuanced sentiments, such as anger, joy, or surprise. This powerful tool has the potential to unlock valuable insights from vast amounts of unstructured text data, which enables informed decision-making and effective communication in various domains, including education. Recent advances in sentiment analysis have leveraged the power of deep neural networks, particularly general-purpose Large Language Models (LLMs) trained on extensive labeled datasets. However, real-world applications frequently encounter challenges related to the availability of large, high-quality labeled data and the computational resources necessary for training such models. This research addresses these challenges by investigating effective strategies for utilizing LLMs in scenarios with limited data and computational resources. Specifically, this study explores three techniques: zero-shot learning, <i>N</i>-shot learning and fine-tuning. By evaluating these methods, this research aims to demonstrate the feasibility of employing general-purpose LLMs for sentiment analysis within educational contexts even when access to computational resources and labeled data is limited. The findings of this study reveal that different adaptation methods lead to significantly different LLM performance outcomes.

Large language models (LLMs) demonstrate impressive capabilities across a range of natural language processing (NLP) tasks. However, they are highly sensitive to prompt design, which significantly affects their ability to align outputs with user intent. Poorly crafted prompts can result in misleading or irrelevant responses. Nevertheless, selecting the most effective prompt from several candidates remains an open challenge. Despite the growing importance of prompt engineering, there is no comprehensive framework to systematically evaluate prompts across multiple dimensions, such as similarity, performance, efficiency, and consistency, particularly in scenarios where performance can be traded off against computational cost or consistency. In this study, we propose a novel scoring framework to evaluate handcrafted prompts across four essential dimensions: Similarity, performance, efficiency (measured by latency, input tokens, and output tokens), and consistency. Considering Arabic, a relatively low-resource, morphologically rich language, as a case study, we evaluated this framework on six diverse text classification tasks: Dialect identification, sentiment analysis, offensive language detection, stance detection, emotion detection, and sarcasm detection. Our methodology assesses prompts across multiple LLMs (GPT-4o mini, LLaMA, ALLAM, and Claude 3.5 Haiku), providing valuable insights into model-specific and task-specific performance patterns. Results demonstrate that no single prompt universally excels across all dimensions; rather, optimal prompts vary based on specific task requirements and evaluation priorities. The proposed framework enables the identification of the most effective prompts for each application context while revealing important trade-offs between performance metrics. By addressing the unique challenges of Arabic NLP, this research not only advances prompt engineering for underrepresented languages but also provides a systematic and adaptable methodology for prompt evaluation that can enhance LLM performance across a range of linguistic contexts, diverse domains, tasks, and various model architectures.

Background and Aims Large language models (LLMs) have gained significant attention in the field of natural language processing (NLP), marking a shift from traditional techniques like Term Frequency-Inverse Document Frequency (TF-IDF). We developed a traditional NLP model to predict arteriovenous fistula (AVF) failure within next 30 days using clinical notes. The goal of this analysis was to investigate whether LLMs would outperform traditional NLP techniques, specifically in the context of predicting AVF failure within the next 30 days using clinical notes. Method We defined AVF failure as the change in status from active to permanently unusable status or temporarily unusable status. We used data from a large kidney care network from January 2021 to December 2021. Two models were created using LLMs and traditional TF-IDF technique. We used “distilbert-base-uncased”, a distilled version of BERT base model [1], and compared its performance with traditional TF-IDF-based NLP techniques. The dataset was randomly divided into 60% training, 20% validation and 20% test dataset. The test data, comprising of unseen patients’ data was used to evaluate the performance of the model. Both models were evaluated using metrics such as area under the receiver operating curve (AUROC), accuracy, sensitivity, and specificity. Results The incidence of 30 days AVF failure rate was 2.3% in the population. Both LLMs and traditional showed similar overall performance as summarized in Table 1. Notably, LLMs showed marginally better performance in certain evaluation metrics. Both models had same AUROC of 0.64 on test data. The accuracy and balanced accuracy for LLMs were 72.9% and 59.7%, respectively, compared to 70.9% and 59.6% for the traditional TF-IDF approach. In terms of specificity, LLMs scored 73.2%, slightly higher than the 71.2% observed for traditional NLP methods. However, LLMs had a lower sensitivity of 46.1% compared to 48% for traditional NLP. However, it is worth noting that training on LLMs took considerably longer than TF-IDF. Moreover, it also used higher computational resources such as utilization of graphics processing units (GPU) instances in cloud-based services, leading to higher cost. Conclusion In our study, we discovered that advanced LLMs perform comparably to traditional TF-IDF modeling techniques in predicting the failure of AVF. Both models demonstrated identical AUROC. While specificity was higher in LLMs compared to traditional NLP, sensitivity was higher in traditional NLP compared to LLMs. LLM was fine-tuned with a limited dataset, which could have influenced its performance to be similar to that of traditional NLP methods. This finding suggests that while LLMs may excel in certain scenarios, such as performing in-depth sentiment analysis of patient data for complex tasks, their effectiveness is highly dependent on the specific use case. It is crucial to weigh the benefits against the resources required for LLMs, as they can be significantly more resource-intensive and costly compared to traditional TF-IDF methods. This highlights the importance of a use-case-driven approach in selecting the appropriate NLP technique for healthcare applications.

BackgroundComplementary therapies are being increasingly used by cancer survivors. As a channel for customers to share their feelings, outcomes, and perceived knowledge about the products purchased from e-commerce platforms, Amazon consumer reviews are a valuable real-world data source for understanding cancer survivorship care needs.ObjectiveIn this study, we aimed to highlight the potential of using Amazon consumer reviews as a novel source for identifying cancer survivorship care needs, particularly related to symptom self-management. Specifically, we present a publicly available, manually annotated corpus derived from Amazon reviews of health-related products and develop baseline natural language processing models using deep learning and large language model (LLM) to demonstrate the usability of this dataset.MethodsWe preprocessed the Amazon review dataset to identify sentences with cancer mentions through a rule-based method and conducted content analysis including text feature analysis, sentiment analysis, topic modeling, cancer type, and symptom association analysis. We then designed an annotation guideline, targeting survivorship-relevant constructs. A total of 159 reviews were annotated, and baseline models were developed based on deep learning and large language model (LLM) for named entity recognition and text classification tasks.ResultsA total of 4703 sentences containing positive cancer mentions were identified, drawn from 3349 reviews associated with 2589 distinct products. The identified topics through topic modeling revealed meaningful insights into cancer symptom management and survivorship experiences. Examples included discussions of green tea use during chemotherapy, cancer prevention strategies, and product recommendations for breast cancer. Top 15 symptoms in reviews were also identified, with pain being the most frequent symptom, followed by inflammation, fatigue, etc. The annotation labels were designed to capture cancer types, indicated symptoms, and symptom management outcomes. The resulting annotation corpus contains 2067 labels from 159 Amazon reviews. It is publicly accessible, together with the annotation guideline through the Open Health Natural Language Processing (OHNLP) GitHub. Our baseline model, Bert-base-cased, achieved the highest weighted average F1-score, that is, 66.92%, for named entity recognition, and LLM gpt4-1106-preview-chat achieved the highest F1-score for text classification tasks, that is, 66.67% for “Harmful outcome,” 88.46% for “Favorable outcome” and 73.33% for “Ambiguous outcome.”ConclusionsOur results demonstrate the potential of Amazon consumer reviews as a novel data source for identifying persistent symptoms, concerns, and self-management strategies among cancer survivors. This corpus, along with the baseline natural language processing models developed for named entity recognition and text classification, lays the groundwork for future methodological advancements in cancer survivorship research. Importantly, insights from this study could be evaluated against established clinical guidelines for symptom management in cancer survivorship care. By revealing the feasibility of using consumer-generated data for mining survivorship-related experiences, this study offers a promising foundation for future research and argumentation analysis aimed at improving long-term outcomes and support for cancer survivors.

DeepSeek has emerged as a prominent representative of China’s domestic large language models(LLMs), attracting widespread public attention and generating diverse discussions on social media since its release. To systematically examine public sentiment and thematic concerns surrounding DeepSeek, this study analyzes 86,008 Weibo posts collected between January 19 and March 22, 2025. After data cleaning, 59,679 valid entries were retained. A mixed-method approach was adopted, combining Latent Dirichlet Allocation (LDA) for topic modeling and a fine-tuned BERT-based classifier for sentiment analysis. The BERT model, fine-tuned on 10,000 hand-labeled posts, achieved an accuracy of 82% and was subsequently applied to the remaining unlabeled data. Topic modeling revealed nine major themes in public discourse, including chip innovation and industrial transformation, smart device ecosystem, and China–US technological competition. Sentiment analysis showed that 46.3% of the posts expressed a positive attitude, reflecting public endorsement of DeepSeek’s domestic innovation and application potential, although concerns about information security, technical feasibility, and international political dynamics were also evident. Additionally, major events significantly influenced emotional fluctuations, with peaks in public attention observed on January 27 and March 19. These findings offer empirical insights for policymakers, developers, and the broader public into public acceptance, key concerns, and potential risks associated with emerging AI technologies, and support the formulation of informed strategies for technology governance, user engagement, and product optimization.

Sentiment analysis, a critical subfield of natural language processing, is widely employed to decipher the emotions andopinions expressed in textual data. With the growing prevalence of emojis in digital communication, understanding theircontribution alongside textual information has become paramount for comprehensive sentiment classification. This paperproposes a novel hybrid deep learning model for sentiment analysis that effectively integrates advanced feature fusion andattention mechanisms to address the challenges of analyzing multimodal data. By combining BERT-based textual embeddingsand pre-trained emoji embeddings, the model captures nuanced semantic and emotional information. A self-attentionmechanism further enhances the representation by identifying long-range dependencies and contextual relationships betweentext and emojis. The model was evaluated on the Sentiment140 dataset, achieving state-of-the-art performance with anaccuracy of 91.7%, an F1-score of 93.6%, and an AUC of 96.4%, outperforming existing models such as BERT-LSTM andRoBERTa-GRU. This superior performance demonstrates the effectiveness of multimodal fusion in sentiment classification,particularly for social media data where emojis play a significant role in emotional expression. The proposed architecture alsoshows strong generalizability, offering robust performance across diverse datasets. While computational complexity is a notedchallenge, future research could explore optimization techniques to improve efficiency without compromising accuracy. Thiswork highlights the potential of hybrid models to advance sentiment analysis by bridging the gap between textual and visual-emotional communication, setting a foundation for more comprehensive multimodal understanding in natural languageprocessing tasks.

Since interpretable feature-engineered pipelines, natural language processing (NLP) has developed into deep neural and transformer-based architectures, and finally, large language models (LLMs) that can generalize their results across tasks. This development has significantly enhanced the subjective language comprehension, such as sentiment, emotion, sarcasm, humor, stance, metaphor, intent and aesthetic judgments, and increased the requirements of explainability in high-stakes areas. Already trained transformers (e.g., encoder-only and decoder-only versions) and LLMs like BERT and GPT have established powerful baselines on text classification and zero/few-shot subjectivity modeling, but the opaqueness of these models drives language-based rationalization (extractive/abstractive natural-language justifications) and feature-attribution algorithms (e.g., LIME, Integrated Gradients, SHAP). The four strands brought together by this survey include: (i) sentiment-analysis methods and datasets (e.g., IMDb, Sentiment140, Twitter Airline, SemEval), (ii) transformer-era text classification, (iii) LLMs on subjective language, and (iv) rationalization of explainable NLP. We have provided a unified taxonomy of subjective tasks and have analyzed model families, starting with classical ML up to transformers/LLMs; have collected major data sets and benchmarks; and have systematized explainability methods, in particular rationalization, and their evaluations and classify open problems dataset bias and annotation ambiguity, constraints of faithfulness and explanatory comprehensibility, evaluation bias, compute cost, and ethical risks. The article will facilitate the formation of a coherent foundation of explainable, credible subjective NLP based on the functions of transformer/LLM and analyze feature extraction techniques methodically based on LLM. We have trained a Machine Learning classifier using 70% of the training data and 30% of the testing data. Based on our results, we find that the proposed ML based technique gives enhanced performance, with an improved accuracy of 99% in the UCI-ML reviews dataset and 96% in the Twitter Kaggle dataset. This study underscores the paramount significance of feature extraction in sentiment analysis, endowing pragmatic insights to elevate model performance and steer future research pursuits.

Cross-category recommendation systems aim to suggest items to users from various categories, enhancing the user experience by broadening their exposure to diverse products or services. Large Language Models (LLMs) offer a promising avenue for improving the effectiveness of such systems. LLMs, trained on vast datasets, excel at understanding complex relationships between different categories of products by analysing patterns in user interactions, preferences, and item descriptions. By leveraging natural language understanding, these models can identify similarities and connections between disparate product categories, enabling more personalized and relevant cross-category recommendations. This approach goes beyond traditional recommendation algorithms that typically focus on a single category or rely heavily on collaborative filtering techniques. Instead, LLMs can process and integrate information from multiple domains, making it easier to suggest complementary products from different categories that users may not have initially considered. Additionally, LLMs can interpret contextual information, such as user reviews and sentiment analysis, to further refine recommendations based on the user's unique preferences and behavioural patterns. The potential of LLMs in cross-category recommendations opens new possibilities for e-commerce, entertainment, and other industries where personalized experiences are key to user engagement and satisfaction. As these models continue to evolve, they offer the promise of more intelligent and comprehensive recommendation systems that can better meet the needs of users by offering diverse and relevant suggestions across categories. These abstract outlines the key role of LLMs in advancing cross-category recommendation systems, highlighting their capacity to improve personalization and broaden the scope of recommendations.

The urgent pursuit of net-zero emissions presents a critical challenge for modern societies, necessitating a speedup of transformative shifts across sectors to mitigate climate change. Predicting trends and drivers in the integration of energy technologies is essential to addressing this challenge, as it informs policy decisions, strategic investments, and the deployment of innovative solutions crucial for transitioning to a sustainable energy future. Despite the importance of accurate forecasting, current methods remain limited, especially in leveraging the vast, unlabelled energy literature available. However, with the advent of large language models (LLMs), the ability to interpret and extract insights from extensive textual data has significantly advanced. Sentiment analysis, in particular, has just emerged as a vital tool for detecting scientific opinions from the energy literature, which can be harnessed to forecast energy trends. This study introduces a novel multi-agent framework, EnergyEval, to evaluate the sentiment and factuality of the energy literature. The core novelty of the multi-agent framework is found to be the use of heterogeneous energy-specialised roles with different LLMs. This investigation, using both multiple persona agents and different LLMs, provides a bespoke collaboration mechanism for multi-agent debate (MAD). In addition, we believe our approach can extend across the energy industry, where deep application of MAD is yet to be exploited. We apply EnergyEval to the case of UK offshore wind literature, assessing its predictive performance. Our findings indicate that the sentiment predicted by the EnergyEval effectively aligns with observed trends in increasing the installed capacity and reductions in Levelised Cost of Energy (LCOE). It also helps us to identify key drivers in offshore wind development. The advantage of employing a multi-agent LLM debate team allows us to achieve competitive accuracy compared to single-LLM-based methods, while significantly reducing computational costs. Overall, the results highlight the potential of EnergyEval as a robust tool for forecasting technology developments in the pursuit of net-zero emissions.

The proliferation of social media has made Sentiment Analysis an essential tool for understanding user opinions, particularly in underrepresented language variants such as Mexican Spanish. Recent advances in Large Language Models have made effective sentiment analysis through in-context learning techniques, reducing the need for supervised training. This study compares the performance of zero and few-shot with traditional fine-tuning approaches of tourism-related texts in Mexican Spanish. Two annotated datasets from the REST-MEX 2022 and 2023 shared tasks were used for this purpose. Results show that fine-tuning, particularly with the MarIA model, achieves the best overall performance. However, modern LLMs that use in-context learning strategies, such as Mixtral 8x7B for zero-shot and Mistral 7B for few-shot, demonstrate strong potential in low-resource settings by closely approximating the accuracy of fine-tuned models, suggesting that in-context learning is a viable alternative to fine-tuning for sentiment analysis in Mexican Spanish when labeled data is limited. These approaches can enable intelligent, data-driven digital services with applications in tourism platforms and urban information systems that enhance user experience and trust in large-scale socio-technical ecosystems.

Sentiment analysis has emerged as a powerful tool in financial markets, offering the ability to harness market participants' collective wisdom and emotions to make more informed investment decisions. This research paper presents a comprehensive literature review on sentiment analysis models in predicting stock prices, focusing on the Bank Nifty Index, a critical benchmark in the Indian financial sector. The review begins by elucidating the significance of sentiment analysis in stock price prediction, highlighting its pivotal role in capturing market sentiment, investor emotions, and their impact on asset valuations. A detailed overview of sentiment sources, encompassing financial news, social media, reports, and other relevant data streams, provides insight into the rich tapestry of information that sentiment analysis can leverage. The paper delves into various sentiment analysis approaches, from traditional lexicon-based and rule-based methods to cutting-edge machine learning and deep learning techniques. It elucidates the intricate process of integrating sentiment analysis with stock price prediction models, exploring feature concatenation, time series integration, and weighted data to incorporate sentiment insights effectively. Evaluation metrics crucial for assessing the performance of both sentiment analysis models and stock price prediction models are thoroughly discussed. It concludes by highlighting the substantial potential of sentiment analysis in augmenting stock price prediction, offering investors and traders valuable tools to navigate the complexities of financial markets.