arXiv subjects

Saptarshi Ghosh

Publications and source records attributed to Saptarshi Ghosh.

At least 73 records · Page 4Linked to original sources

arXiv2023

A Bi-Level Stochastic Game Model for PMU Placement in Power Grid with Cybersecurity Risks

Phasor measurement units (PMUs) provide accurate and high-fidelity measurements in order to monitor the state of the power grid and support various control and planning tasks. However, PMUs have a high installation cost prohibiting their massive deployment. Minimizing the number of installed PMUs needs to be achieved while also maintaining full observability of the network. At the same time, data integrity attacks on PMU measurements can cause mislead power system control and operation routines. In this paper, a bi-level stochastic non-cooperative game-based placement model is proposed for PMU allocation in the presence of cyber-attack risks. In the first level, the protection of individual PMU placed in a network is addressed, while considering the interaction between the grid operator and the attacker with respective resource constraints. In the second level, the attacker observes the placement of the PMUs and compromises them, with the aim of maximizing the state estimation error and reducing the observability of the network. The proposed technique is deployed in the IEEE-9 bus test system. The results demonstrate a 9% reduction in the cost incurred by the power grid operator for deploying PMUs while considering cyber-risks.

eess.SY↗

arXiv2023

Few-Shot Visual Question Generation: A Novel Task and Benchmark Datasets

Generating natural language questions from visual scenes, known as Visual Question Generation (VQG), has been explored in the recent past where large amounts of meticulously labeled data provide the training corpus. However, in practice, it is not uncommon to have only a few images with question annotations corresponding to a few types of answers. In this paper, we propose a new and challenging Few-Shot Visual Question Generation (FS-VQG) task and provide a comprehensive benchmark to it. Specifically, we evaluate various existing VQG approaches as well as popular few-shot solutions based on meta-learning and self-supervised strategies for the FS-VQG task. We conduct experiments on two popular existing datasets VQG and Visual7w. In addition, we have also cleaned and extended the VQG dataset for use in a few-shot scenario, with additional image-question pairs as well as additional answer categories. We call this new dataset VQG-23. Several important findings emerge from our experiments, that shed light on the limits of current models in few-shot vision and language generation tasks. We find that trivially extending existing VQG approaches with transfer learning or meta-learning may not be enough to tackle the inherent challenges in few-shot VQG. We believe that this work will contribute to accelerating the progress in few-shot learning research.

cs.CV↗

arXiv2022

Analyzing Regrettable Communications on Twitter: Characterizing Deleted Tweets and Their Authors

Over 500 million tweets are posted in Twitter each day, out of which about 11% tweets are deleted by the users posting them. This phenomenon of widespread deletion of tweets leads to a number of questions: what kind of content posted by users makes them want to delete them later? %Are all users equally active in deleting their tweets or Are users of certain predispositions more likely to post regrettable tweets, deleting them later? In this paper we provide a detailed characterization of tweets posted and then later deleted by their authors. We collected tweets from over 200 thousand Twitter users during a period of four weeks. Our characterization shows significant personality differences between users who delete their tweets and those who do not. We find that users who delete their tweets are more likely to be extroverted and neurotic while being less conscientious. Also, we find that deleted tweets while containing less information and being less conversational, contain significant indications of regrettable content. Since users of online communication do not have instant social cues (like listener's body language) to gauge the impact of their words, they are often delayed in employing repair strategies. Finally, we build a classifier which takes textual, contextual, as well as user features to predict if a tweet will be deleted or not. The classifier achieves a F1-score of 0.78 and the precision increases when we consider response features of the tweets.

cs.SI↗

arXiv2022

What You Like: Generating Explainable Topical Recommendations for Twitter Using Social Annotations

With over 500 million tweets posted per day, in Twitter, it is difficult for Twitter users to discover interesting content from the deluge of uninteresting posts. In this work, we present a novel, explainable, topical recommendation system, that utilizes social annotations, to help Twitter users discover tweets, on topics of their interest. A major challenge in using traditional rating dependent recommendation systems, like collaborative filtering and content based systems, in high volume social networks is that, due to attention scarcity most items do not get any ratings. Additionally, the fact that most Twitter users are passive consumers, with 44% users never tweeting, makes it very difficult to use user ratings for generating recommendations. Further, a key challenge in developing recommendation systems is that in many cases users reject relevant recommendations if they are totally unfamiliar with the recommended item. Providing a suitable explanation, for why the item is recommended, significantly improves the acceptability of recommendation. By virtue of being a topical recommendation system our method is able to present simple topical explanations for the generated recommendations. Comparisons with state-of-the-art matrix factorization based collaborative filtering, content based and social recommendations demonstrate the efficacy of the proposed approach.

cs.IR↗

arXiv2022

Thickness mapping and layer number identification of exfoliated van der Waals materials by Fourier imaging micro-ellipsometry

As properties of mono- to few layers of exfoliated van der Waals heterostructures are heavily dependent on their thicknesses, accurate thickness measurement becomes imperative in their study. Commonly used atomic force microscopy and Raman spectroscopy techniques may be invasive and produce inconclusive results. Alternatively, spectroscopic ellipsometry is limited by tens-of-microns lateral resolution and/or low data acquisition rates, inhibiting its utilization for micro-scale exfoliated flakes. In this work, we demonstrate a Fourier imaging spectroscopic micro-ellipsometer with sub-5 microns lateral resolution along with fast data acquisition rate and present angstrom-level accurate and consistent thickness mapping on mono-, bi- and trilayers of graphene, hexagonal boron nitride and transition metal dichalcogenide (MoS2, WS2, MoSe2, WSe2) flakes. We show that the optical microscope integrated ellipsometer can also map minute thickness variations over a micro-scale flake. In addition, our system addresses the pertinent issue of identifying monolayer thick hBN.

physics.optics↗

arXiv2022

CAVES: A Dataset to facilitate Explainable Classification and Summarization of Concerns towards COVID Vaccines

Convincing people to get vaccinated against COVID-19 is a key societal challenge in the present times. As a first step towards this goal, many prior works have relied on social media analysis to understand the specific concerns that people have towards these vaccines, such as potential side-effects, ineffectiveness, political factors, and so on. Though there are datasets that broadly classify social media posts into Anti-vax and Pro-Vax labels, there is no dataset (to our knowledge) that labels social media posts according to the specific anti-vaccine concerns mentioned in the posts. In this paper, we have curated CAVES, the first large-scale dataset containing about 10k COVID-19 anti-vaccine tweets labelled into various specific anti-vaccine concerns in a multi-label setting. This is also the first multi-label classification dataset that provides explanations for each of the labels. Additionally, the dataset also provides class-wise summaries of all the tweets. We also perform preliminary experiments on the dataset and show that this is a very challenging dataset for multi-label explainable classification and tweet summarization, as is evident by the moderate scores achieved by some state-of-the-art models. Our dataset and codes are available at: https://github.com/sohampoddar26/caves-data

cs.CL↗

arXiv2022

ECTSum: A New Benchmark Dataset For Bullet Point Summarization of Long Earnings Call Transcripts

Despite tremendous progress in automatic summarization, state-of-the-art methods are predominantly trained to excel in summarizing short newswire articles, or documents with strong layout biases such as scientific articles or government reports. Efficient techniques to summarize financial documents, including facts and figures, have largely been unexplored, majorly due to the unavailability of suitable datasets. In this work, we present ECTSum, a new dataset with transcripts of earnings calls (ECTs), hosted by publicly traded companies, as documents, and short experts-written telegram-style bullet point summaries derived from corresponding Reuters articles. ECTs are long unstructured documents without any prescribed length limit or format. We benchmark our dataset with state-of-the-art summarizers across various metrics evaluating the content quality and factual consistency of the generated summaries. Finally, we present a simple-yet-effective approach, ECT-BPS, to generate a set of bullet points that precisely capture the important facts discussed in the calls.

cs.CL↗

arXiv2022

Legal Case Document Summarization: Extractive and Abstractive Methods and their Evaluation

Summarization of legal case judgement documents is a challenging problem in Legal NLP. However, not much analyses exist on how different families of summarization models (e.g., extractive vs. abstractive) perform when applied to legal case documents. This question is particularly important since many recent transformer-based abstractive summarization models have restrictions on the number of input tokens, and legal documents are known to be very long. Also, it is an open question on how best to evaluate legal case document summarization systems. In this paper, we carry out extensive experiments with several extractive and abstractive summarization methods (both supervised and unsupervised) over three legal summarization datasets that we have developed. Our analyses, that includes evaluation by law practitioners, lead to several interesting insights on legal summarization in specific and long document summarization in general.

cs.CL↗

arXiv2022

Legal Case Document Similarity: You Need Both Network and Text

Estimating the similarity between two legal case documents is an important and challenging problem, having various downstream applications such as prior-case retrieval and citation recommendation. There are two broad approaches for the task -- citation network-based and text-based. Prior citation network-based approaches consider citations only to prior-cases (also called precedents) (PCNet). This approach misses important signals inherent in Statutes (written laws of a jurisdiction). In this work, we propose Hier-SPCNet that augments PCNet with a heterogeneous network of Statutes. We incorporate domain knowledge for legal document similarity into Hier-SPCNet, thereby obtaining state-of-the-art results for network-based legal document similarity. Both textual and network similarity provide important signals for legal case similarity; but till now, only trivial attempts have been made to unify the two signals. In this work, we apply several methods for combining textual and network information for estimating legal case similarity. We perform extensive experiments over legal case documents from the Indian judiciary, where the gold standard similarity between document-pairs is judged by law experts from two reputed Law institutes in India. Our experiments establish that our proposed network-based methods significantly improve the correlation with domain experts' opinion when compared to the existing methods for network-based legal document similarity. Our best-performing combination method (that combines network-based and text-based similarity) improves the correlation with domain experts' opinion by 11.8% over the best text-based method and 20.6\% over the best network-based method. We also establish that our best-performing method can be used to recommend / retrieve citable and similar cases for a source (query) case, which are well appreciated by legal experts.

cs.IR↗

arXiv2022

Two-stream Plasma Instability as a Potential Mechanism for Particle Escape from the Venusian Ionosphere

In this work we investigate the possibility of two-stream instability in the Venusian atmosphere to lead to momentum transfer to subsequent escape of Hydrogen and Oxygen ions from the ionosphere. We employ the hydrodynamic model and obtain the linear dispersion relation from which the two-stream instability is studied. Further the interaction of solar wind with the ions of Venus ionosphere from which the instability sets in, has been studied with the data from ASPERA-4 of Venus Express (VEX). The data supports the fact that the two-stream instability can provide sufficient energy to accelerate ions to escape velocity of the planet.

physics.plasm-ph↗

arXiv2022

FaiRIR: Mitigating Exposure Bias from Related Item Recommendations in Two-Sided Platforms

Related Item Recommendations (RIRs) are ubiquitous in most online platforms today, including e-commerce and content streaming sites. These recommendations not only help users compare items related to a given item, but also play a major role in bringing traffic to individual items, thus deciding the exposure that different items receive. With a growing number of people depending on such platforms to earn their livelihood, it is important to understand whether different items are receiving their desired exposure. To this end, our experiments on multiple real-world RIR datasets reveal that the existing RIR algorithms often result in very skewed exposure distribution of items, and the quality of items is not a plausible explanation for such skew in exposure. To mitigate this exposure bias, we introduce multiple flexible interventions (FaiRIR) in the RIR pipeline. We instantiate these mechanisms with two well-known algorithms for constructing related item recommendations -- rating-SVD and item2vec -- and show on real-world data that our mechanisms allow for a fine-grained control on the exposure distribution, often at a small or no cost in terms of recommendation quality, measured in terms of relatedness and user satisfaction.

cs.IR↗

arXiv2022

Alexa, in you, I trust! Fairness and Interpretability Issues in E-commerce Search through Smart Speakers

In traditional (desktop) e-commerce search, a customer issues a specific query and the system returns a ranked list of products in order of relevance to the query. An increasingly popular alternative in e-commerce search is to issue a voice-query to a smart speaker (e.g., Amazon Echo) powered by a voice assistant (VA, e.g., Alexa). In this situation, the VA usually spells out the details of only one product, an explanation citing the reason for its selection, and a default action of adding the product to the customer's cart. This reduced autonomy of the customer in the choice of a product during voice-search makes it necessary for a VA to be far more responsible and trustworthy in its explanation and default action. In this paper, we ask whether the explanation presented for a product selection by the Alexa VA installed on an Amazon Echo device is consistent with human understanding as well as with the observations on other traditional mediums (e.g., desktop ecommerce search). Through a user survey, we find that in 81% cases the interpretation of 'a top result' by the users is different from that of Alexa. While investigating for the fairness of the default action, we observe that over a set of as many as 1000 queries, in nearly 68% cases, there exist one or more products which are more relevant (as per Amazon's own desktop search results) than the product chosen by Alexa. Finally, we conducted a survey over 30 queries for which the Alexa-selected product was different from the top desktop search result, and observed that in nearly 73% cases, the participants preferred the top desktop search result as opposed to the product chosen by Alexa. Our results raise several concerns and necessitates more discussions around the related fairness and interpretability issues of VAs for e-commerce search.

cs.HC↗

arXiv2021

LeSICiN: A Heterogeneous Graph-based Approach for Automatic Legal Statute Identification from Indian Legal Documents

The task of Legal Statute Identification (LSI) aims to identify the legal statutes that are relevant to a given description of Facts or evidence of a legal case. Existing methods only utilize the textual content of Facts and legal articles to guide such a task. However, the citation network among case documents and legal statutes is a rich source of additional information, which is not considered by existing models. In this work, we take the first step towards utilising both the text and the legal citation network for the LSI task. We curate a large novel dataset for this task, including Facts of cases from several major Indian Courts of Law, and statutes from the Indian Penal Code (IPC). Modeling the statutes and training documents as a heterogeneous graph, our proposed model LeSICiN can learn rich textual and graphical features, and can also tune itself to correlate these features. Thereafter, the model can be used to inductively predict links between test documents (new nodes whose graphical features are not available to the model) and statutes (existing nodes). Extensive experiments on the dataset show that our model comfortably outperforms several state-of-the-art baselines, by exploiting the graphical structure along with textual features. The dataset and our codes are available at https://github.com/Law-AI/LeSICiN.

cs.CL↗

arXiv2021

Winds of Change: Impact of COVID-19 on Vaccine-related Opinions of Twitter users

Administering COVID-19 vaccines at a societal scale has been deemed as the most appropriate way to defend against the COVID-19 pandemic. This global vaccination drive naturally fueled a possibility of Pro-Vaxxers and Anti-Vaxxers strongly expressing their supports and concerns regarding the vaccines on social media platforms. Understanding this online discourse is crucial for policy makers. This understanding is likely to impact the success of vaccination drives and might even impact the final outcome of our fight against the pandemic. The goal of this work is to improve this understanding using the lens of Twitter-discourse data. We first develop a classifier that categorizes users according to their vaccine-related stance with high precision (97%). Using this method we detect and investigate specific user-groups who posted about vaccines in pre-COVID and COVID times. Specifically, we identify distinct topics that these users talk about, and investigate how vaccine-related discourse has changed between pre-COVID times and COVID times. Finally, for the first time, we investigate the change of vaccine-related stances in Twitter users and shed light on potential reasons for such changes in stance. Our dataset and classifier are available at https://github.com/sohampoddar26/covid-vax-stance.

cs.SI↗

arXiv2021

Incorporating Domain Knowledge for Extractive Summarization of Legal Case Documents

Automatic summarization of legal case documents is an important and practical challenge. Apart from many domain-independent text summarization algorithms that can be used for this purpose, several algorithms have been developed specifically for summarizing legal case documents. However, most of the existing algorithms do not systematically incorporate domain knowledge that specifies what information should ideally be present in a legal case document summary. To address this gap, we propose an unsupervised summarization algorithm DELSumm which is designed to systematically incorporate guidelines from legal experts into an optimization setup. We conduct detailed experiments over case documents from the Indian Supreme Court. The experiments show that our proposed unsupervised method outperforms several strong baselines in terms of ROUGE scores, including both general summarization algorithms and legal-specific ones. In fact, though our proposed algorithm is unsupervised, it outperforms several supervised summarization models that are trained over thousands of document-summary pairs.

cs.CL↗

arXiv2021

Machine Learning assisted Chimera and Solitary states in Networks

Chimera and Solitary states have captivated scientists and engineers due to their peculiar dynamical states corresponding to the co-existence of coherent and incoherent dynamical evolution in coupled units in various natural and artificial systems. It has been further demonstrated that such states can be engineered in systems of coupled oscillators by the suitable implementation of communication delays. Here, using supervised machine learning, we predict (a) the precise value of delay which is sufficient for engineering chimera and solitary states for a given set of system parameters, as well as (b) the intensity of incoherence for such engineered states. The results are demonstrated for two different examples consisting of single layer and multi layer networks. First, the chimera states (solitary states) are engineered by establishing delays in the neighboring links of a node (the interlayer links) in a 2-D lattice (multiplex network) of oscillators. Then, different machine learning classifiers, KNN, SVM and MLP-Neural Network are employed by feeding the data obtained from the network models. Once a machine learning model is trained using a limited amount of data, it makes predictions for a given unknown systems parameter values. Testing accuracy, sensitivity, and specificity analysis reveal that MLP-NN classifier is better suited than Knn or SVM classifier for the predictions of parameters values for engineered chimera and solitary states. The technique provides an easy methodology to predict critical delay values as well as the intensity of incoherence for designing an experimental setup to create solitary and chimera states.

nlin.AO↗

arXiv2021

Fairness for Whom? Understanding the Reader's Perception of Fairness in Text Summarization

With the surge in user-generated textual information, there has been a recent increase in the use of summarization algorithms for providing an overview of the extensive content. Traditional metrics for evaluation of these algorithms (e.g. ROUGE scores) rely on matching algorithmic summaries to human-generated ones. However, it has been shown that when the textual contents are heterogeneous, e.g., when they come from different socially salient groups, most existing summarization algorithms represent the social groups very differently compared to their distribution in the original data. To mitigate such adverse impacts, some fairness-preserving summarization algorithms have also been proposed. All of these studies have considered normative notions of fairness from the perspective of writers of the contents, neglecting the readers' perceptions of the underlying fairness notions. To bridge this gap, in this work, we study the interplay between the fairness notions and how readers perceive them in textual summaries. Through our experiments, we show that reader's perception of fairness is often context-sensitive. Moreover, standard ROUGE evaluation metrics are unable to quantify the perceived (un)fairness of the summaries. To this end, we propose a human-in-the-loop metric and an automated graph-based methodology to quantify the perceived bias in textual summaries. We demonstrate their utility by quantifying the (un)fairness of several summaries of heterogeneous socio-political microblog datasets.

cs.IR↗

arXiv2021

When the Umpire is also a Player: Bias in Private Label Product Recommendations on E-commerce Marketplaces

Algorithmic recommendations mediate interactions between millions of customers and products (in turn, their producers and sellers) on large e-commerce marketplaces like Amazon. In recent years, the producers and sellers have raised concerns about the fairness of black-box recommendation algorithms deployed on these marketplaces. Many complaints are centered around marketplaces biasing the algorithms to preferentially favor their own `private label' products over competitors. These concerns are exacerbated as marketplaces increasingly de-emphasize or replace `organic' recommendations with ad-driven `sponsored' recommendations, which include their own private labels. While these concerns have been covered in popular press and have spawned regulatory investigations, to our knowledge, there has not been any public audit of these marketplace algorithms. In this study, we bridge this gap by performing an end-to-end systematic audit of related item recommendations on Amazon. We propose a network-centric framework to quantify and compare the biases across organic and sponsored related item recommendations. Along a number of our proposed bias measures, we find that the sponsored recommendations are significantly more biased toward Amazon private label products compared to organic recommendations. While our findings are primarily interesting to producers and sellers on Amazon, our proposed bias measures are generally useful for measuring link formation bias in any social or content networks.

cs.CY↗