Sentiment analysis is the process of determining the sentiment polarity (positivity, neutrality or negativity) of the text. As online markets have become more popular over the past decades, online retailers and merchants are asking their buyers to share their opinions about the products they have purchased. As a result, millions of reviews are generated daily, making it difficult to make a good decision about whether a consumer should buy a product. Analyzing these enormous concepts is difficult and time-consuming for product manufacturers. Deep learning is the current research interest in Natural language processing. In the proposed model, Skip-gram architecture is used for better feature extraction of semantic and contextual information of words. LSTM (long short-term memory) is used in the proposed model for understanding complex patterns in textual data. To improve the performance of the LSTM, weight parameters are optimized by the adaptive particle Swarm Optimization algorithm. Extensive experiments were conducted on four datasets proved that our proposed APSO-LSTM model secured higher accuracy over the classical methods such as traditional LSTM, ANN, and SVM. According to simulation results, the proposed model is outperforming other existing models in different metrics.

情感分析是确定文本情感极性（积极、中立或消极）的过程。随着过去几十年在线市场变得越来越流行，在线零售商和商家要求他们的买家分享他们对所购买产品的看法。因此，每天都会产生数百万条评论，因此很难就消费者是否应该购买产品做出正确的决定。对于产品制造商来说，分析这些庞大的概念既困难又耗时。深度学习是当前自然语言处理的研究兴趣。在所提出的模型中，Skip-gram 架构用于更好地提取单词的语义和上下文信息的特征。LSTM（长短期记忆）用于所提出的模型中，用于理解文本数据中的复杂模式。为了提高 LSTM 的性能，权重参数通过自适应粒子群优化算法进行优化。在四个数据集上进行的大量实验证明，我们提出的 APSO-LSTM 模型比传统的 LSTM、ANN 和 SVM 等经典方法具有更高的准确性。根据仿真结果，所提出的模型在不同的指标上优于其他现有模型。和支持向量机。根据仿真结果，所提出的模型在不同的指标上优于其他现有模型。和支持向量机。根据仿真结果，所提出的模型在不同的指标上优于其他现有模型。