High-tech services in smart cities, ubiquity of smart phones, and proliferation of social media platforms have enabled social sensing, either through direct human observers or through humans as sensor carriers and operators, such as through the use of smart phones, cameras, etc. We performed a sentiment analysis (SA) and mined public opinion on the civil services and policing authority in a smart city. The establishment of high-tech policing in Lahore, Pakistan, known as the Punjab Safe Cities Authority (PSCA), Lahore, along with integrated command and control centers and various equipments, such as 8,000 cameras, monitoring sensors, etc., has resulted in a requirement for its performance evaluation and social media–enabled opinion mining to determine the broader impact on communities. Social sensing of civil services has been enabled through the presence of the PSCA on Facebook, Twitter, YouTube, and Web TV. The SA of the local civil services is not possible without taking into account the local language. In this article, we utilize machine learning techniques to perform multi-class SA of public opinion on policing authority and the provided civil services in both the local languages Urdu and English. The support vector machine provides the highest performance multi-classification accuracy of 86.87% for positive, negative, and neutral sentiments. The temporal sentiments are determined over time from January 2020 to July 2021, with an overall positive sentiment of 62.40% and a negative sentiment of 13.51%, which shows high satisfaction of policing authority and the provided civil services.

Modern means of communication, economic crises, and political decisions play imperative roles in reshaping political and administrative systems throughout the world. Twitter, a micro-blogging website, has gained paramount importance in terms of public opinion-sharing. Manual intelligence of law enforcement agencies (i.e., in changing situations) cannot cope in real time. Thus, to address this problem, we built an alert system for government authorities in the province of Punjab, Pakistan. The alert system gathers real-time data from Twitter in English and Roman Urdu about forthcoming gatherings (protests, demonstrations, assemblies, rallies, sit-ins, marches, etc.). To determine public sentiment regarding upcoming anti-government gatherings (protests, demonstrations, assemblies, rallies, sit-ins, marches, etc.), the alert system determines the polarity of tweets. Using keywords, the system provides information for future gatherings by extracting the entities like date, time, and location from Twitter data obtained in real time. Our system was trained and tested with different machine learning (ML) algorithms, such as random forest (RF), decision tree (DT), support vector machine (SVM), multinomial naïve Bayes (MNB), and Gaussian naïve Bayes (GNB), along with two vectorization techniques, i.e., term frequency–inverse document frequency (TFIDF) and count vectorization. Moreover, this paper compares the accuracy results of sentiment analysis (SA) of Twitter data by applying supervised machine learning (ML) algorithms. In our research experiment, we used two data sets, i.e., a small data set of 1000 tweets and a large data set of 4000 tweets. Results showed that RF along with count vectorization performed best for the small data set with an accuracy of 82%; with the large data set, MNB along with count vectorization outperformed all other classifiers with an accuracy of 75%. Additionally, language models, e.g., bigram and trigram, were used to generate the word clouds of positive and negative words to visualize the most frequently used words.

IoT devices produce a vast amount of data ranging from personal to sensitive information. Usually, these devices remain connected to the internet so protecting the information produced by them is crucial. Since most of the IoT devices are resource-constrained, they must be supported with light-weight encryption standards to protect information. Recent research has used the concept of context awareness to select the most suitable data encryption standard based on the device resources along with the required information confidentiality level. However, to effectively use the context information, it is required to be organized explicitly while considering the dynamic nature of IoT systems. In this regard, ontology-based systems effectively reduce the volume of manual work while recommending solutions. Currently, these systems cannot work with precision due to multiple uncertain factors of IoT sensory data,. To overcome this challenge, this research proposes a fuzzy ontology-based context-aware system to protect IoT device information with the help of an encryption algorithm that considers device capabilities and user priorities regarding the data confidentiality. In-order to automate the recommendation process, Semantic Web Rule Language (SWRL) rules and fuzzy logic are used, whereas, Description Logic and RDF Query Language is used to evaluate the results. The evaluation results confirm that the proposed method can produce results according to human perception by significantly increasing the accuracy of prediction.