Maria Despoina Siampou

Maria Despoina Siampou

Auto-generated profile, based on publicly available information

Maria Despoina Siampou

University of Southern California, Los Angeles, California, USA

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 November 2023
Verified
Conference Paper
Open access
AHArash Hajisafi
Arash Hajisafi
HLHaowen Lin
Haowen Lin
SSSina Shaham
Sina Shaham
HHHaoji Hu
Haoji Hu
MSMaria Despoina Siampou
Maria Despoina Siampou
CSCyrus Shahabi
Cyrus Shahabi
Forecasting the number of visits to Points-of-Interest (POI) in an urban area is critical for planning and decision making in various application domains, from urban planning and transportation management to public health and social studies. Although this forecasting problem can be formulated as a multivariate time-series forecasting task, current approaches cannot fully exploit the ever-changing multi-context correlations among POIs. Therefore, we propose Busyness Graph Neural Network (BysGNN), a temporal graph neural network designed to learn and uncover the underlying multi-context correlations between POIs for accurate visit forecasting. Unlike other approaches where only time-series data is used to learn a dynamic graph, BysGNN utilizes all contextual information and time-series data to learn an accurate dynamic graph representation. By incorporating all contextual, temporal, and spatial signals, we observe a significant improvement in our forecasting accuracy over state-of-the-art forecasting models in our experiments with real-world datasets across the United States.

3 November 2025
Verified
Conference Paper
Open access
BABita Azarijoo
Bita Azarijoo
MSMaria Despoina Siampou
Maria Despoina Siampou
JKJohn Krumm
John Krumm
CSCyrus Shahabi
Cyrus Shahabi
Abnormal human mobility patterns often signal disruptions, emergencies, or health-related risks, making their detection critical for applications in public safety, urban monitoring, and healthcare. Existing approaches for human mobility anomaly detection typically focus on either identifying visits to unusual places or overall deviations from individual- and population-level norms at the agent-level. However, these methods often (1) overlook fine-grained temporal anomalies, and (2) lack interpretability, as they do not reveal which specific spatiotemporal components of a visit contribute to its anomalous nature. To overcome these limitations, we present ICAD (Interpretable Component-wise Anomaly Detection), a self-supervised autoregressive model that detects both spatial and temporal anomalies by modeling deviations in an individual’s visit-level mobility behavior. ICAD is trained on normal visit sequences using a next-visit prediction objective to learn the distribution of visits under regular conditions. At inference, it computes component-wise anomaly scores for each visit by measuring relative divergence from the learned distribution of normal behavior. Specifically, ICAD proposes a top-k deviation metric for discrete spatial anomalies and introduces a novel relative mode-based scoring function for detecting temporal anomalies in continuous time. Experiments on a large scale synthetic human mobility dataset show that ICAD outperforms prior methods in both visit-level and agent-level anomaly detection. For reproducability purposes, the source code is accessible at https://github.com/USC-InfoLab/ICAD.

13 November 2023
Verified
Conference Paper
Open access
MSMaria Despoina Siampou
Maria Despoina Siampou
GPGeorge Papadakis
George Papadakis
NMNikos Mamoulis
Nikos Mamoulis
MKManolis Koubarakis
Manolis Koubarakis
Geospatial Interlinking constitutes a crucial data integration task that associates pairs of geometries with topological relations. Its high computational cost, though, scales poorly to voluminous datasets. Progressive methods were recently proposed to reduce this cost by sacrificing recall to an affordable extent. They operate in a learning-free manner that relies on mere heuristics, which can be conservative (i.e., retaining too many unrelated pairs) or aggressive (i.e., discarding too many related pairs). In this work, we extend them with Supervised Scheduling, a quick and principled way of defining the processing order of the candidate geometry pairs that are likely to be topologically related, based on their classification probability. Our approach leverages generic features with low extraction cost but high discriminatory power. We integrate Supervised Scheduling into a progressive end-to-end algorithm that automatically labels the required training instances at a low computational cost. Thorough experiments verify the high performance and robustness of our features as well as the limited size of the training set that suffices for learning an accurate classification model. Our experiments also verify the superior performance of our approach in comparison to existing learning-free ones over five real, large datasets.

3 November 2025
Verified
Conference Paper
Open access
MSMaria Despoina Siampou
Maria Despoina Siampou
SHShang-Ling Hsu
Shang-Ling Hsu
SCShushman Choudhury
Shushman Choudhury
NANeha Arora
Neha Arora
CSCyrus Shahabi
Cyrus Shahabi
Recent advances in large foundation models (FMs) have enabled learning general-purpose representations in natural language, vision, and audio. Yet geospatial artificial intelligence (GeoAI) still lacks widely adopted foundation models that generalize across tasks that require joint reasoning over geospatial objects and human mobility. Such tasks are crucial as mobility, along with satellite imagery, street view, and text, is a core modality for understanding the physical world. We argue that a key bottleneck is the absence of unified, general-purpose, and transferable representations for geospatially embedded objects (GEOs). Such objects include points, polylines, and polygons in geographic space, enriched with semantic context and critical for geospatial reasoning. Much current GeoAI research compares GEOs to tokens in language models, where patterns of human movement and spatiotemporal interactions yield contextual meaning similar to patterns of words in text. However, modeling GEOs introduces challenges fundamentally different from language, including spatial continuity, variable scale and resolution, temporal dynamics, and data sparsity. Moreover, privacy constraints and global variation in mobility further complicates modeling and generalization. This paper formalizes these challenges, identifies key representational gaps, and outlines research directions for building foundation models that learn behavior-informed, transferable representations of GEOs from large-scale human mobility data, as well as static contextual information such as points of interest, object shapes and spatio-temporal semantics.