Browsing by Author "Chatterjee, S."

Now showing 1 - 10 of 10

An uncertainty-aware decision support system: Integrating text narratives and conformal prediction for trustworthy accident code classification
(Institution of Chemical Engineers, 2025) Kumar, A.; Senapati, A.; Upadhyay, R.; Chatterjee, S.; Bhattacherjee, A.; Samanta, B.
It is imperative to assign accident classification codes to the Mine Safety and Health Administration (MSHA) accident data for effective data analysis and risk assessment. Although trained personnel are capable of performing this task, the manual process is both time-consuming and resource-intensive. Automating the classification process with machine learning (ML) algorithms promises to expedite code assignment. However, ML predictions typically lack uncertainty metrics. This study proposes an uncertainty-aware hierarchical classification framework that assists human experts in efficiently and accurately assigning accident codes. Several text representation techniques combined with different ML algorithms were employed within a hierarchical architecture to assign classification codes. Low-frequency codes were consolidated into a single category, with a primary classifier distinguishing between these and a secondary classifier further classifying the grouped categories. Regularized Adaptive Prediction Sets (RAPS) was integrated to quantify uncertainty. Highly confident predictions yielding single-class sets were automatically classified, whereas multi-class sets were flagged for manual review. Primary Classifier with XGBoost with word2vec text representation achieved the best performance, with 95.12 % coverage, 37.02 % single-class prediction sets at 96.11 % accuracy, and an average prediction set size of 2.39. Whereas the secondary classifier, a logistic regression model with TF-IDF representation, yielded 96.19 % coverage, an average set size of 1.80, and 53.66 % single-class prediction sets with 98.90 % accuracy. Additionally, sensitivity analysis determined that a 95 % coverage guarantee offers the best trade-off between prediction set size and coverage. The framework effectively integrates conformal prediction to quantify uncertainty and aid human experts in improving the decision-making process in safety management. Although the framework is broadly applicable across different sectors, it needs to be retrained on domain-specific data for effective use. © 2025 The Institution of Chemical Engineers
Distributed data aggregation in wireless sensor network: With peer verification
(2016) Chatterjee, S.; Pais, A.R.; Saurabh, S.
Data aggregation in wireless sensor network is implemented to reduce the communication overhead and to reduce bandwidth utilization. Data confidentiality requires the sensor node to transmit the data in a secure manner so that the adversary is unable to read the data or transmit false data even if it compromises some of the sensor nodes or aggregation node. In this paper a distributed aggregation protocol using homomorphic trapdoor permutation is proposed. This protocol distributes the responsibility of key generation, aggregation and verification to different nodes to reduce the overall power consumption of the sensor network. The peer verification scheme is also proposed as a part of the protocol. Peer verification ensures the authentication of the data and sender node in the network, by at least k peer nodes. Security of the proposed protocol is analyzed against passive and active adversary model. Copyright � 2016 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.
Distributed data aggregation in wireless sensor network: With peer verification
(SciTePress, 2016) Chatterjee, S.; Pais, A.R.; Saurabh, S.
Data aggregation in wireless sensor network is implemented to reduce the communication overhead and to reduce bandwidth utilization. Data confidentiality requires the sensor node to transmit the data in a secure manner so that the adversary is unable to read the data or transmit false data even if it compromises some of the sensor nodes or aggregation node. In this paper a distributed aggregation protocol using homomorphic trapdoor permutation is proposed. This protocol distributes the responsibility of key generation, aggregation and verification to different nodes to reduce the overall power consumption of the sensor network. The peer verification scheme is also proposed as a part of the protocol. Peer verification ensures the authentication of the data and sender node in the network, by at least k peer nodes. Security of the proposed protocol is analyzed against passive and active adversary model. Â© Â© 2016 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.
Efficient key management in IoT using mobile aggregator
(2016) Saurabh, S.; Pais, A.R.; Chatterjee, S.
Managing keys in Internet of Things (IoT) is challenging. With this proposed work we are trying to address an efficient key management protocol for specific application based scenario which enforces secure connectivity of devices and minimizes node capture attacks. There are a number of protocols that have been enforced and implemented for wireless sensor networks (WSN) and internet-enabled devices. We propose a protocol with mobility interface using combinatorial designs for key management in IoT devices. Mobile devices follow a dedicated path to collect data securely from installed devices in the network. We also compare our work with existing protocols and few mobility models. � Springer Nature Singapore Pte Ltd. 2016.
Efficient key management in IoT using mobile aggregator
(Springer Verlag service@springer.de, 2016) Saurabh, S.; Pais, A.R.; Chatterjee, S.
Managing keys in Internet of Things (IoT) is challenging. With this proposed work we are trying to address an efficient key management protocol for specific application based scenario which enforces secure connectivity of devices and minimizes node capture attacks. There are a number of protocols that have been enforced and implemented for wireless sensor networks (WSN) and internet-enabled devices. We propose a protocol with mobility interface using combinatorial designs for key management in IoT devices. Mobile devices follow a dedicated path to collect data securely from installed devices in the network. We also compare our work with existing protocols and few mobility models. Â© Springer Nature Singapore Pte Ltd. 2016.
Hierarchical secret sharing scheme using parts of speech of English grammar
(Inderscience Publishers, 2017) Chatterjee, S.; Koolagudi, S.G.
In this paper, a model to share secret information in conjunctive and disjunctive hierarchical access structure using obfuscation is proposed. Indistinguishability Obfuscation is achieved with a context free grammar (CFG) as a mimic function. Obfuscation is used to maintain confidentiality of the message in the presence of a dishonest distributor who is curious to know the secret. A new way to effectively reduce the size of the share is also achieved in this model. First, mimic function is used to convert the statistical profile of the message to a random distribution of words from a chosen paragraph. The frequency distribution of different parts of speech (PoS) components of the obfuscated string is used to build a model for distribution of shares to n people based on the responsibility of the person in a hierarchy. Sharing the information and reconstruction of the original message is also shown. It is also shown that the obfuscation is secure against chosen plain text attack. © © 2017 Inderscience Enterprises Ltd.
Searchable encryption through obfuscation and multi-cloud search
(2016) Chatterjee, S.; Bhattacharjee, S.; Chandrasekaran, K.
In this paper different searchable encryption techniques are studied along with some implementation of searchable encryption in a cloud environment. Symmetric searchable encryption is implemented through trapdoor function to selectively expose keyword for search. A new method of achieving searchable encryption in the random oracle model is proposed through one way indistinguishability obfuscation. Indistinguishability obfuscation is achievable through mimicry function and one way cryptographic hash function. Security of the model is also analyzed with non adaptable indistinguishability security. Using this method an efficient idea of multi-cloud search environment is also proposed. In multi cloud searchable encryption several independent data providers collaborate in a federated way to provide search in their data through different independent cloud service provider revealing only keywords associated with the data. � 2015 IEEE.
Searchable encryption through obfuscation and multi-cloud search
(Institute of Electrical and Electronics Engineers Inc., 2016) Chatterjee, S.; Bhattacharjee, S.; Chandrasekaran, K.
In this paper different searchable encryption techniques are studied along with some implementation of searchable encryption in a cloud environment. Symmetric searchable encryption is implemented through trapdoor function to selectively expose keyword for search. A new method of achieving searchable encryption in the random oracle model is proposed through one way indistinguishability obfuscation. Indistinguishability obfuscation is achievable through mimicry function and one way cryptographic hash function. Security of the model is also analyzed with non adaptable indistinguishability security. Using this method an efficient idea of multi-cloud search environment is also proposed. In multi cloud searchable encryption several independent data providers collaborate in a federated way to provide search in their data through different independent cloud service provider revealing only keywords associated with the data. Â© 2015 IEEE.
Studies on the surface oxidation of molten aluminium - Silicon alloys using laser as a probe
(2007) Udupa, K.R.; Chatterjee, S.
An attempt has been made to study for the first time the surface oxidation of molten aluminium-silicon alloy using laser as a probe. A ray of low powered laser was made to fall on freshly prepared surface of molten aluminium-silicon alloy and intensity of the reflected beam was measured using a light detecting resistor (LDR). The change in intensity of reflected laser beam which was due to surface degradation of the molten liquid on oxidation was recorded. Intensity profile which showed variation of intensity (registered as reciprocal of resistance on LDR) with respect to time was generated for pure aluminium melt and molten aluminium-silicon alloy. The intensity profile revealed three regions indicating different mechanisms of oxidation in each stage. It was found that alloying addition influences the nature of intensity profile. Copyright � 2007 MS&T'07�.
Studies on the surface oxidation of molten aluminium - Silicon alloys using laser as a probe
(Materials Science and Technology, 2007) Udupa, K.R.; Chatterjee, S.
An attempt has been made to study for the first time the surface oxidation of molten aluminium-silicon alloy using laser as a probe. A ray of low powered laser was made to fall on freshly prepared surface of molten aluminium-silicon alloy and intensity of the reflected beam was measured using a light detecting resistor (LDR). The change in intensity of reflected laser beam which was due to surface degradation of the molten liquid on oxidation was recorded. Intensity profile which showed variation of intensity (registered as reciprocal of resistance on LDR) with respect to time was generated for pure aluminium melt and molten aluminium-silicon alloy. The intensity profile revealed three regions indicating different mechanisms of oxidation in each stage. It was found that alloying addition influences the nature of intensity profile. Copyright Â© 2007 MS&T'07Â®.