arXiv subjects

Manish Singh

Publications and source records attributed to Manish Singh.

15 recordsLinked to original sources

arXiv2025

Leveraging Spatiotemporal Graph Neural Networks for Multi-Store Sales Forecasting

This work evaluates the effectiveness of spatiotemporal Graph Neural Networks (GNNs) for multi-store retail sales forecasting and compares their performance against ARIMA, LSTM, and XGBoost baselines. Using weekly sales data from 45 Walmart stores, we construct a relational forecasting framework that models inter-store dependencies through a learned adaptive graph. The proposed STGNN predicts log-differenced sales and reconstructs final values through a residual path, enabling stable training and improved generalisation. Experiments show that STGNN achieves the lowest overall forecasting error, outperforming all baselines in Normalised Total Absolute Error, P90 MAPE, and variance of MAPE across stores. Analysis of the learned adjacency matrix reveals meaningful functional store clusters and high-influence nodes that emerge without geographic metadata. These results demonstrate that relational structure significantly improves forecast quality in interconnected retail environments and establishes STGNNs as a robust modelling choice for multi-store demand prediction.

cs.LG↗

arXiv2025

Options-Aware Dense Retrieval for Multiple-Choice query Answering

Long-context multiple-choice question answering tasks require robust reasoning over extensive text sources. Since most of the pre-trained transformer models are restricted to processing only a few hundred words at a time, successful completion of such tasks often relies on the identification of evidence spans, such as sentences, that provide supporting evidence for selecting the correct answer. Prior research in this domain has predominantly utilized pre-trained dense retrieval models, given the absence of supervision to fine-tune the retrieval process. This paper proposes a novel method called Options Aware Dense Retrieval (OADR) to address these challenges. ORDA uses an innovative approach to fine-tuning retrieval by leveraging query-options embeddings, which aim to mimic the embeddings of the oracle query (i.e., the query paired with the correct answer) for enhanced identification of supporting evidence. Through experiments conducted on the QuALITY benchmark dataset, we demonstrate that our proposed model surpasses existing baselines in terms of performance and accuracy.

cs.IR↗

arXiv2024

Beyond Spatio-Temporal Representations: Evolving Fourier Transform for Temporal Graphs

We present the Evolving Graph Fourier Transform (EFT), the first invertible spectral transform that captures evolving representations on temporal graphs. We motivate our work by the inadequacy of existing methods for capturing the evolving graph spectra, which are also computationally expensive due to the temporal aspect along with the graph vertex domain. We view the problem as an optimization over the Laplacian of the continuous time dynamic graph. Additionally, we propose pseudo-spectrum relaxations that decompose the transformation process, making it highly computationally efficient. The EFT method adeptly captures the evolving graph's structural and positional properties, making it effective for downstream tasks on evolving graphs. Hence, as a reference implementation, we develop a simple neural model induced with EFT for capturing evolving graph spectra. We empirically validate our theoretical findings on a number of large-scale and standard temporal graph benchmarks and demonstrate that our model achieves state-of-the-art performance.

cs.LG↗

arXiv2023

Initial boundary value problem for 1D scalar balance laws with strictly convex flux

A Lax-Oleinik type explicit formula for 1D scalar balance laws has been recently obtained for the pure initial value problem by Adimurthi et al. in [1]. In this article, by introducing a suitable boundary functional, we establish a Lax-Oleinik type formula for the initial boundary value problem. For the pure initial value problem, the solution for the corresponding Hamilton-Jacobi equation turns out to be the minimizer of a functional on the set of curves known as h-curves. In the present situation, part of the h-curve joining any two points in the quarter plane may cross the boundary $x = 0.$ This phenomenon breaks the simplicity of the minimization process through the boundary functional compared to the case of conservation laws. Moreover, this complicates the verification of the boundary condition in the sense of Bardos, le Roux, and Nedelec [2]. To verify the boundary condition, the boundary points are classified into three types depending on the structure of the minimizers at those points. Finally, by introducing characteristic triangles, we construct generalized characteristics and show that the explicit solution is entropy admissible.

math.AP↗

arXiv2023

Can Persistent Homology provide an efficient alternative for Evaluation of Knowledge Graph Completion Methods?

In this paper we present a novel method, $\textit{Knowledge Persistence}$ ($\mathcal{KP}$), for faster evaluation of Knowledge Graph (KG) completion approaches. Current ranking-based evaluation is quadratic in the size of the KG, leading to long evaluation times and consequently a high carbon footprint. $\mathcal{KP}$ addresses this by representing the topology of the KG completion methods through the lens of topological data analysis, concretely using persistent homology. The characteristics of persistent homology allow $\mathcal{KP}$ to evaluate the quality of the KG completion looking only at a fraction of the data. Experimental results on standard datasets show that the proposed metric is highly correlated with ranking metrics (Hits@N, MR, MRR). Performance evaluation shows that $\mathcal{KP}$ is computationally efficient: In some cases, the evaluation time (validation+test) of a KG completion method has been reduced from 18 hours (using Hits@10) to 27 seconds (using $\mathcal{KP}$), and on average (across methods & data) reduces the evaluation time (validation+test) by $\approx$ $\textbf{99.96}\%$.

cs.LG↗

arXiv2022

Learnable Spectral Wavelets on Dynamic Graphs to Capture Global Interactions

Learning on evolving(dynamic) graphs has caught the attention of researchers as static methods exhibit limited performance in this setting. The existing methods for dynamic graphs learn spatial features by local neighborhood aggregation, which essentially only captures the low pass signals and local interactions. In this work, we go beyond current approaches to incorporate global features for effectively learning representations of a dynamically evolving graph. We propose to do so by capturing the spectrum of the dynamic graph. Since static methods to learn the graph spectrum would not consider the history of the evolution of the spectrum as the graph evolves with time, we propose a novel approach to learn the graph wavelets to capture this evolving spectra. Further, we propose a framework that integrates the dynamically captured spectra in the form of these learnable wavelets into spatial features for incorporating local and global interactions. Experiments on eight standard datasets show that our method significantly outperforms related methods on various tasks for dynamic graphs.

cs.LG↗

arXiv2022

Defects and Metric Anomalies in Föppl-von Kármán Surfaces

A general framework is developed to study the deformation and stress response in F{ö}ppl-von K{á}rm{á}n shallow shells for a given distribution of defects, such as dislocations, disclinations, and interstitials, and metric anomalies, such as thermal and growth strains. The theory includes dislocations and disclinations whose defect lines can both pierce the two-dimensional surface and lie within the surface. An essential aspect of the theory is the derivation of strain incompatibility relations for stretching and bending strains with incompatibility sources in terms of various defect and metric anomaly densities. The incompatibility relations are combined with balance laws and constitutive assumptions to obtain the inhomogeneous F{ö}ppl-von K{á}rm{á}n equations for shallow shells. Several boundary value problems are posed, and solved numerically, by first considering only dislocations and then disclinations coupled with growth strains.

cond-mat.soft↗

arXiv2021

Positive disclination in a thin elastic sheet with boundary

An isolated positive wedge disclination deforms an initially flat elastic sheet into a perfect cone when the sheet is of infinite extent and is elastically inextensible. The latter requires the elastic stretching strains to be vanishingly small. In this paper, rigorous analytical and numerical results are obtained for the disclination induced deformed shape and stress field of a bounded F{ö}ppl-von K{á}rm{á}n elastic sheet with finite extensibility, while emphasising the deviations from the perfect cone solution. In particular, the Gaussian curvature field is no longer localised as a Dirac singularity at the defect location whenever elastic extensibility is allowed and is necessarily negative in large regions away from the defect. The stress field, similarly, has no Dirac singularity in the presence of elastic extensibility. However, with increasing Young's modulus of the sheet, while keeping the bending modulus and the domain size fixed, both of these fields tend to develop a Dirac singularity. Noticeably, in this limiting behaviour, inextensibility eludes the bounded elastic sheet due to persisting regions of non-trivial Gaussian curvature away from the defect. Other results in the paper include studying the effect of specific boundary conditions (free, simply supported, or partially clamped) on the Gaussian curvature field away from the defect and on the buckling transition from the flat to a conical solution.

cond-mat.soft↗

arXiv2020

Chemical segregation in Ge2Sb2Te5 thin films during in-situ heating

Germanium antimony telluride has been the most used and studied phase-change material for electronic memory due to its suitable crystallization temperature, amorphous to crystalline resistance contrast, and stability of the amorphous phase. In this work, the segregation of Ge in a Ge2Sb2Te5 film of 30 nm thickness during heating inside the transmission electron microscope was observed and characterized. The Ge2Sb2Te5 film was deposited using sputtering on a Protochips Fusion holder and left uncapped in atmosphere for about four months. Oxygen incorporated within the film played a significant role in the chemical segregation observed which resulted in amorphous Ge-O grain boundaries and Sb and Te rich crystalline domains. Such composition changes can occur when the phase-change material interfaces insulating oxide layers in an integrated device and would significantly impact its electrical and thermal properties.

cond-mat.mtrl-sci↗

arXiv2019

Fabrication of p+ contact by thermally induced solid state regrowth of Al on p-type Ge crystal

Formation of p+ contact on Germanium is important for applications in diode detector and other electronic devices. In this work, thermally deposited Al on Ge crystal is annealed at 350oC followed by slow cooling for solid-state regrowth of Al-Ge p+ contact on Ge. Depth profile analysis by secondary ion mass spectrometry (SIMS) is carried out to investigate the occurrence of Al and Ge along the depth of the regrown Al-Ge layer. Evidence of regrowth is observed due to inter-diffusion of both Ge and Al across the layers although Ge diffusion in Al layer is found to be more prevalent. Thickness of the evaporated Al layer is varied to understand the diffusion profile of Al, Ge and estimate the depth of Al incorporation in Ge crystal underneath. Hall measurement at different depth of Al-Ge regrown layer reveals that Al impurity induces p+ doping in p-type Ge and its concentration gradually reduces towards the Ge crystal. Top surface of the Al-Ge layer exhibits lowest sheet resistance that varies with the thickness of the as deposited Al layer.

cond-mat.mtrl-sci↗

arXiv2019

Using Social Media for Word-of-Mouth Marketing

Nowadays online social networks are used extensively for personal and commercial purposes. This widespread popularity makes them an ideal platform for advertisements. Social media can be used for both direct and word-of-mouth (WoM) marketing. Although WoM marketing is considered more effective and it requires less advertisement cost, it is currently being under-utilized. To do WoM marketing, we need to identify a set of people who can use their authoritative position in social network to promote a given product. In this paper, we show how to do WoM marketing in Facebook group, which is a question answer type of social network. We also present concept of reinforced WoM marketing, where multiple authorities can together promote a product to increase the effectiveness of marketing. We perform our experiments on Facebook group dataset consisting of 0.3 million messages and 10 million user reactions.

cs.SI↗

arXiv2019

Toward Maximizing the Visibility of Content in Social Media Brand Pages: A Temporal Analysis

A large amount of content is generated everyday in social media. One of the main goals of content creators is to spread their information to a large audience. There are many factors that affect information spread, such as posting time, location, type of information, number of social connections, etc. In this paper, we look at the problem of finding the best posting time(s) to get high content visibility. The posting time is derived taking other factors into account, such as location, type of information, etc. In this paper, we do our analysis over Facebook pages. We propose six posting schedules that can be used for individual pages or group of pages with similar audience reaction profile. We perform our experiment on a Facebook pages dataset containing 0.3 million posts, 10 million audience reactions. Our best posting schedule can lead to seven times more number of audience reactions compared to the average number of audience reactions that users would get without following any optimized posting schedule. We also present some interesting audience reaction patterns that we obtained through daily, weekly and monthly audience reaction analysis.

cs.SI↗

arXiv2019

Sentiment Dynamics in Social Media News Channels

Social media is currently one of the most important means of news communication. Since people are consuming a large fraction of their daily news through social media, most of the traditional news channels are using social media to catch the attention of users. Each news channel has its own strategies to attract more users. In this paper, we analyze how the news channels use sentiment to garner users' attention in social media. We compare the sentiment of social media news posts of television, radio and print media, to show the differences in the ways these channels cover the news. We also analyze users' reactions and opinion sentiment on news posts with different sentiments. We perform our experiments on a dataset extracted from Facebook Pages of five popular news channels. Our dataset contains 0.15 million news posts and 1.13 billion users reactions. The results of our experiments show that the sentiment of user opinion has a strong correlation with the sentiment of the news post and the type of information source. Our study also illustrates the differences among the social media news channels of different types of news sources.

cs.SI↗

arXiv2019

FastGRNN: A Fast, Accurate, Stable and Tiny Kilobyte Sized Gated Recurrent Neural Network

This paper develops the FastRNN and FastGRNN algorithms to address the twin RNN limitations of inaccurate training and inefficient prediction. Previous approaches have improved accuracy at the expense of prediction costs making them infeasible for resource-constrained and real-time applications. Unitary RNNs have increased accuracy somewhat by restricting the range of the state transition matrix's singular values but have also increased the model size as they require a larger number of hidden units to make up for the loss in expressive power. Gated RNNs have obtained state-of-the-art accuracies by adding extra parameters thereby resulting in even larger models. FastRNN addresses these limitations by adding a residual connection that does not constrain the range of the singular values explicitly and has only two extra scalar parameters. FastGRNN then extends the residual connection to a gate by reusing the RNN matrices to match state-of-the-art gated RNN accuracies but with a 2-4x smaller model. Enforcing FastGRNN's matrices to be low-rank, sparse and quantized resulted in accurate models that could be up to 35x smaller than leading gated and unitary RNNs. This allowed FastGRNN to accurately recognize the "Hey Cortana" wakeword with a 1 KB model and to be deployed on severely resource-constrained IoT microcontrollers too tiny to store other RNN models. FastGRNN's code is available at https://github.com/Microsoft/EdgeML/.

cs.LG↗

arXiv2017

Efficient Twitter Sentiment Classification using Subjective Distant Supervision

As microblogging services like Twitter are becoming more and more influential in today's globalised world, its facets like sentiment analysis are being extensively studied. We are no longer constrained by our own opinion. Others opinions and sentiments play a huge role in shaping our perspective. In this paper, we build on previous works on Twitter sentiment analysis using Distant Supervision. The existing approach requires huge computation resource for analysing large number of tweets. In this paper, we propose techniques to speed up the computation process for sentiment analysis. We use tweet subjectivity to select the right training samples. We also introduce the concept of EFWS (Effective Word Score) of a tweet that is derived from polarity scores of frequently used words, which is an additional heuristic that can be used to speed up the sentiment classification with standard machine learning algorithms. We performed our experiments using 1.6 million tweets. Experimental evaluations show that our proposed technique is more efficient and has higher accuracy compared to previously proposed methods. We achieve overall accuracies of around 80% (EFWS heuristic gives an accuracy around 85%) on a training dataset of 100K tweets, which is half the size of the dataset used for the baseline model. The accuracy of our proposed model is 2-3% higher than the baseline model, and the model effectively trains at twice the speed of the baseline model.

cs.SI↗