Privacy preserving sentiment analysis on multiple edge data streams with Apache NiFi

Abstract

Sentiment analysis, also known as opinion mining, plays a big role in both private and public sector Business Intelligence (BI); it attempts to improve public and customer experience. Nevertheless, de-identified sentiment scores from public social media posts can compromise individual privacy due to their vulnerability to record linkage attacks. Established privacy-preserving methods like k-anonymity, l-diversity and tcloseness are offline models exclusively designed for data at rest. Recently, a number of online anonymization algorithms (CASTLE, SKY, SWAF) have been proposed to complement the functional requirements of streaming applications, but without open-source implementation. In this paper, we present a reusable Apache NiFi dataflow that buffers tweets from multiple edge devices and performs anonymized sentiment analysis in real-time, using randomization. The solution can be easily adapted to suit different scenarios, enabling researchers to deploy custom anonymization algorithms.

A Novel Edge Architecture and Solution for Detecting Concept Drift in Smart Environments

“…Smart cities' promise of digitalisation necessitates additional deployments of IoT or a multitude of the aforementioned technologies, posing enormous challenges [9,26,37]. Examples of such challenges include ensuring the security and privacy of smart city platforms ( [38], data heterogeneity issues, sensors and their connectivity, big data analytics, scalability, reliability, and the fault-tolerance of services [27,37]. There have been several surveys on these subjects [33,37,39].…”

Towards Collective Sentiment Analysis in IoT-Enabled Scenarios

“…However, it leverages social media content, which means that the user feedback is explicitly provided and also there is no privacy in general. Approaches for (both offline and online) privacy-preserving processing of social media content are also possible, leveraging anonymisation algorithms [17], such as in [18], where the authors provide a proof-of-concept for SA-aimed anonymisation of edge datastreams. However, though the approach is privacy-preserving, it leverages Twitter content, i.e., text manually provided by users, whereas our approach provides implicit or transparent acquisition of user feedback.…”

"The I-Nergy Reference Architecture for the Provision of Next Generation Energy Services Through Artificial Intelligence"

“…The ingested data are also available to third parties and external stakeholders through a service based on blockchain and smart contracts that secures trusted data sharing and controlled access to data. The main technologies that are used in the Data Management component are Apache NiFi (Pandya et al, 2019), Apache Kafka (Kafka, n.d.) and MongoDB (MongoDB, 2018). Apache NiFi supports powerful directed graphs for data routing, transformation and system mediation logic, and within I-NERGY it facilitates the data ingestion and preprocessing phase.…”