Enhancing machine learning-based sentiment analysis through feature extraction techniques.

A crucial part of sentiment classification is featuring extraction because it involves extracting valuable information from text data, which affects the model's performance. The goal of this paper is to help in selecting a suitable feature extraction method to enhance the performance of sentiment analysis tasks. In order to provide directions for future machine learning and feature extraction research, it is important to analyze and summarize feature extraction techniques methodically from a machine learning standpoint. There are several methods under consideration, including Bag-of-words (BOW), Word2Vector, N-gram, Term Frequency- Inverse Document Frequency (TF-IDF), Hashing Vectorizer (HV), and Global vector for word representation (GloVe). To prove the ability of each feature extractor, we applied it to the Twitter US airlines and Amazon musical instrument reviews datasets. Finally, we trained a random forest classifier using 70% of the training data and 30% of the testing data, enabling us to evaluate and compare the performance using different metrics. Based on our results, we find that the TD-IDF technique demonstrates superior performance, with an accuracy of 99% in the Amazon reviews dataset and 96% in the Twitter US airlines 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.

Genomic Profiling of Rare Undifferentiated Sarcomatoid Subtypes of Pancreatic Carcinomas: In Search of Therapeutic Targets.

PURPOSE: The highly aggressive undifferentiated sarcomatoid carcinoma (USC) subtype of pancreatic ductal adenocarcinoma (PDAC) remains poorly characterized because of its rarity. Previous case reports suggest that immune checkpoint inhibitors could be a promising treatment strategy, but the prevalence of established predictive biomarkers of response is largely unknown. The objective of this study was to leverage comprehensive genomic profiling of USC PDAC tumors to determine the prevalence of biomarkers associated with potential response to targeted therapies. METHODS: USC tumors (n = 20) underwent central pathology review by a board-certified gastrointestinal pathologist to confirm the diagnosis. These samples were compared with non-USC PDAC tumors (N = 5,562). Retrospective analysis of DNA and RNA next-generation sequencing data was performed. RESULTS: USC PDACs were more frequently PD-L1+ by immunohistochemistry than non-USC PDAC (63% v 16%, respectively, P < .001). Furthermore, USC PDAC had an increase in neutrophils (8.99% v 5.55%, P = .005) and dendritic cells (1.08% v 0.00%, q = 0.022) and an increased expression of PDCD1LG2 (4.6% v 1.3%, q = 0.001), PDCD1 (2.0% v 0.8%, q = 0.060), and HAVCR2 (45.9% v 21.7%, q = 0.107) than non-USC PDAC. Similar to non-USC PDAC, KRAS was the most commonly mutated gene (86% v 90%, respectively, P = 1). CONCLUSION: To our knowledge, this work represents the largest molecular analysis of USC tumors to date and showed an increased expression of immune checkpoint genes in USC tumors. These findings provide evidence for further investigation into immune checkpoint inhibitors in USC tumors.

BOLERO-5: a phase II study of everolimus and exemestane combination in Chinese post-menopausal women with ER + /HER2- advanced breast cancer.

BACKGROUND: The global BOLERO-2 trial established the efficacy and safety of combination everolimus (EVE) and exemestane (EXE) in the treatment of estrogen receptor positive (ER +), HER2-, advanced breast cancer (ABC). BOLERO-5 investigated this combination in a Chinese population (NCT03312738). METHODS: BOLERO-5 is a randomized, double-blind, multicenter, placebo controlled, phase II trial comparing EVE (10 mg/day) or placebo (PBO) in combination with EXE (25 mg/day). The primary endpoint was progression-free survival (PFS) per investigator assessment. Secondary endpoints included PFS per blinded independent review committee (BIRC), overall survival (OS), overall response rate (ORR), clinical benefit rate (CBR), pharmacokinetics, and safety. RESULTS: A total of 159 patients were randomized to EVE + EXE (n = 80) or PBO + EXE (n = 79). By investigator assessment, treatment with EVE + EXE prolonged median PFS by 5.4 months (HR 0.52; 90% CI 0.38, 0.71), from 2.0 months (PBO + EXE; 90% CI 1.9, 3.6) to 7.4 months (EVE + EXE; 90% CI 5.5, 9.0). Similar results were observed following assessment by BIRC, with median PFS prolonged by 4.3 months. Treatment with EVE + EXE was also associated with improvements in ORR and CBR. No new safety signals were identified in BOLERO-5, with the incidence of adverse events in Chinese patients consistent with the safety profile of both drugs. CONCLUSION: The efficacy and safety results of BOLERO-5 validate the findings from BOLERO-2, and further support the use of EVE + EXE in Chinese post-menopausal women with ER + , HER2- ABC. NCT03312738, registered 18 October 2017.

Improving sentiment classification using a RoBERTa-based hybrid model.

Introduction: Several attempts have been made to enhance text-based sentiment analysis's performance. The classifiers and word embedding models have been among the most prominent attempts. This work aims to develop a hybrid deep learning approach that combines the advantages of transformer models and sequence models with the elimination of sequence models' shortcomings. Methods: In this paper, we present a hybrid model based on the transformer model and deep learning models to enhance sentiment classification process. Robustly optimized BERT (RoBERTa) was selected for the representative vectors of the input sentences and the Long Short-Term Memory (LSTM) model in conjunction with the Convolutional Neural Networks (CNN) model was used to improve the suggested model's ability to comprehend the semantics and context of each input sentence. We tested the proposed model with two datasets with different topics. The first dataset is a Twitter review of US airlines and the second is the IMDb movie reviews dataset. We propose using word embeddings in conjunction with the SMOTE technique to overcome the challenge of imbalanced classes of the Twitter dataset. Results: With an accuracy of 96.28% on the IMDb reviews dataset and 94.2% on the Twitter reviews dataset, the hybrid model that has been suggested outperforms the standard methods. Discussion: It is clear from these results that the proposed hybrid RoBERTa-(CNN+ LSTM) method is an effective model in sentiment classification.

Primary Pancreatic Liposarcoma: An Unexpected Cause of a Pancreatic Mass.

Liposarcoma is the most common type of soft-tissue sarcoma and typically occurs in the extremities or retroperitoneum. Primary liposarcoma of the pancreas is exceedingly rare, with only 10 cases reported since 1979. We present a patient who was incidentally discovered to have a pancreatic mass on imaging, which was ultimately diagnosed as dedifferentiated pancreatic liposarcoma. We review the clinical and histologic features of pancreatic liposarcoma in this case and in the 10 previously reported cases to increase awareness and knowledge of this rare disease.