6 July 2021 in IEEE Access
Non-Technical Loss (NTL) is a significant concern for many electric supply companies due to the financial impact caused as a result of suspect consumption activities. A range of machine learning classifiers have been tested across multiple synthesized and real datasets to combat NTL. An important characteristic that exists in these datasets is the imbalance distribution of the classes. When the focus is on predicting the minority class of suspect activities, the classifiers’ sensitivity to the class imbalance becomes more important. In this paper, we evaluate the performance of a range of classifiers with under-sampling and over-sampling techniques. The results are compared with the untreated imbalanced dataset. In addition, we compare the performance of the classifiers using penalized classification model. Lastly, the paper presents an exploratory analysis of using different sampling techniques on NTL detection in a real dataset and identify the best performing classifiers. We conclude that logistic regression is the most sensitive to the sampling techniques as the change of its recall is measured around 50% for all sampling techniques. While the random forest is the least sensitive to the sampling technique, the difference in its precision is observed between 1% – 6% for all sampling techniques.
9 January 2023 in IEEE Access
People use web search engines to retrieve information from the world wide web. Search engines maintain query logs to refine retrieved information for personalized web search. The query log includes information of users, both sensitive and non-sensitive. If utilized against an individual, the query log may pose privacy concerns and reveal a lot of information about them. In recent years many private web search (PWS) schemes have been proposed to realize privacy-preserving web search. Although each PWS scheme claims to have unique features for attaining web search privacy (WSP), no study explains which private web search characteristics should be considered when building and utilizing a PWS scheme. There are two objectives of this article. In the first part of the article, we present a novel PWS scheme that uses a proxy-query-based query obfuscation approach. Proxy-query-based query obfuscation is a new study topic in PWS research. It provides an IR facility for retrieving information from web search engines via proxy queries. One clear advantage of the proposed scheme is that users do not issue true queries to search engines for retrieving information. The article’s second objective is to define the characteristics of PWS and analyze modern PWS schemes on these characteristics. We analyzed proposed and modern PWS schemes on the PWS characteristics. The analysis demonstrated only the proposed PWS scheme achieves all characteristics. Existing PWS systems have been discovered vulnerable to WSP attacks since they do not meet all PWS characteristics.
28 April 2026 in IEEE Access
Current watermarking approaches for LLM-generated text typically embed watermarks during the generation process, using a secret key controlled by the LLM. Although this enables verification of LLM-generated content, it may raise copyright and authorship concerns. For instance, when a user relies on an LLM for proofreading, grammar correction, or translation, the model may still embed its own watermark. This can later create disputes regarding ownership, as the text may appear to belong to the AI system rather than the human author. To address this issue, this paper proposes an improved post-hoc watermarking approach that embeds user-controlled watermarks after the text generation is complete. The approach does not require access to the internal architecture of the LLM. It applies a probabilistic watermarking mechanism using a user secret key, and embeds watermark signals across the entire text rather than only on selective terms, as in previous post-hoc approaches. During watermark embedding, the approach regenerates the suffix after each token substitution by interacting with the LLM. This preserves grammatical correctness and contextual coherence, addressing the quality degradation often observed in synonym-based substitution approaches that do not adjust the surrounding text. A BERT-based masked language model is used to ensure that each substitution remains semantically consistent with the original text. Experimental results show that the proposed approach maintains higher similarity and readability than previous post-hoc approaches while producing statistically verifiable watermark signals. The approach provides a practical solution for authorship verification and copyright protection in generative AI, without requiring access to the internal mechanism of the LLM.