Quantifying displacement: an urban expansion consequence via persistent homology
summary
The gist
Population displacement is defined as a housing-related involuntary residential dislocation The results reveal its ability to capture displacement and identify the neighbourhoods and years…
In short
This study uses advanced Topological Data Analysis to quantify housing-related involuntary residential displacement in Madrid from 2004 to 2023. By creating four simultaneous cubical complexes from address change data, the method identifies specific neighborhoods and years affected by movement patterns, revealing complex displacement dynamics not visible through simple data analysis.
Key concepts
- Population Displacement
- Defined as an involuntary residential dislocation related to housing. The study uses this definition to track how people move and where they settle over time in a city.
- Cubical Complexes
- These are 3D structures built from geographical (space) and temporal (time) information of people moving. They allow researchers to visualize movement patterns across different dimensions simultaneously.
- Persistent Homology
- A mathematical tool used to analyze the shape and structure of the cubical complexes. It helps quantify displacement by identifying specific topological features, such as holes, that persist over time.
Terminology used across episodes
This episode discusses
- Quantifying displacement: an urban expansion consequence via persistent homology · Paper Radio
- Understanding U.S. Racial Segregation Through Persistent Homology
- Cubical Ripser: Software for computing persistent homology of image and volume data
- Evaluating Cooling Center Coverage Using Persistent Homology of a Filtered Witness Complex
- A Comparison of Precinct and District Voting Data Using Persistent Homology to Identify Gerrymandering in North Carolina
The paper
Quantifying displacement: an urban expansion consequence via persistent homology · Read on arXiv
RITA RODRÁGUEZ VAZQUEZ, MANUEL CUERNO
CUNEF Universidad
Transcript
Introduction to the show: ident: Computer Science. The day in computer science: software, systems and theory on arXiv.
Tom: We're looking at the paper: "Quantifying displacement: an urban expansion consequence via persistent homology".
Jane: Here's the gist of the paper. Population displacement is defined as a housing-related involuntary residential dislocation The results reveal its ability to capture displacement and identify the neighbourhoods and years affected–patterns not observable from raw address change data This novel methodology uses publicly available address change data and Topological Data Analysis tools to quantify displacement Methodological Novelty The authors introduce a novel tool using publicly available address change data to construct four cubical complexes simultaneously incorporating geographical and temporal information of people moving They analyze this data using Persistent Homology to quantify displacement This approach overcomes limitations in existing studies by not relying on ad-hoc methodologies or city-specific metrics, making the method highly replicable across cities and time periods.
Tom: The construction involves dividing individuals into four groups based on origin and destination, assuming that individuals develop social and practical ties in their area of residence.
Jane: Data Construction The study uses yearly address change data for the city of Madrid, spanning from 2004 until 2023.
Tom: The data is preprocessed by dividing individuals into four groups: ‘stay’, ‘city’, ‘C.
Title and authors: Tom: So, Jane, the paper "Quantifying displacement: an urban expansion consequence via persistent homology" tackles how we actually measure population displacement in cities. It’s about moving beyond just looking at raw address changes because that doesn't give you the full picture of what’s happening over time.
Jane: Right, so it addresses the difficulty of capturing displacement because it’s a gradual process and you only have snapshots of data, which is what they start with on page one. So, how do they actually go about measuring this involuntary residential dislocation?
Tom: Well, they introduce a novel tool using publicly available address change data to construct four cubical complexes at the same time. They incorporate both the geographical location and the temporal information of people moving. It’s a big step because it aims to capture displacement patterns that you just can't see with simple raw address change data.
Lu: What's interesting about their methodology is that they use Persistent Homology to quantify this displacement, which is a Topological Data Analysis tool. This allows them to analyze the data in a way that’s not ad-hoc or city-specific, which should make it more replicable across different cities and time periods.
Meng: Replicability sounds good in theory, but for me, I wonder how robust this construction is when you move it from Madrid to some other city where the administrative boundaries change frequently. What’s the practical cost of building these cubical complexes?
Jane: That's a fair point about administrative changes. They address that in their data preprocessing by detailing how they handle neighborhood boundary changes, specifically noting they merge people moving within a district into all neighborhoods in Vicálvaro from two thousand four to two thousand sixteen and then keep things as is for later periods like district nineteen.
Tom: They also mention how they handle the creation of new administrative units, like neighborhood one hundred eighty-three Ensanche de Vallecas by assigning emigration rates from its predecessor neighborhood. That shows they thought about keeping the map consistent even when the city structure evolves.
Lalam: From my perspective, it seems like their main goal is to overcome the critiques in existing literature that rely on ad-hoc techniques and data, and inadequate time scale, which they tackle by incorporating time as one of the axes in their construction, which is novel both in content and approach <ref:2512.10753#pg3>.
Title and authors: Lu: And they ground their use of persistent homology by using a construction tailored specifically to this problem that fully exploits the strengths of persistent homology. That's a strong methodological claim about why it works for displacement measurement.
Tom: Okay, so if we look at the results from their twenty-year case study in Madrid, what are we actually seeing in terms of what they found <ref:2512.10753#pg1,20-year case study in Madrid>? What do these topological features tell us about where people are moving?
Jane: They show distinctive displacement dynamics in specific neighborhoods and years. For example, they found that neighborhood "twenty-seven Atocha" behaves differently from its surroundings throughout the entire twenty-year period, which is a key pattern to watch for <ref:2512.10753#pg2>.
Meng: So it’s not just one big number; it’s localized behavior. But what about the other groups they defined, like 'stay' or 'city'? How does their analysis differentiate between people who are just staying put versus those who are actually displaced?
Lu: They use different topological features to differentiate them. Features in H0 correspond to enclaves of people moving to a particular destination, and highly persistent H0 features are relevant enclaves for that specific group.
Tom: And for the 'stay' group, they found higher dimensional topological features, specifically one- and two-holes, suggesting displacement. They point out that H2 features in the 'stay' group indicate significant population displacement occurred in the central city area since two thousand thirteen coinciding with the post-two thousand eight crisis recovery <ref:2512.10753#pg2>.
Jane: That links it directly to a known historical event, which is important context for understanding how these patterns emerge. So, if we look at the other groups, what do they show?
Tom: The 'city' group exhibited a larger number of connected components than the others, suggesting that this group is more dispersed across both space and time. On the flip side, H0 features in the 'outside' group were more indicative of displacement, with neighborhood "one hundred forty-one Pavones (East)" appearing to experience high pressure.
Meng: So there’s a contrast: one group is spread out, another shows localized pressure points in specific areas. That difference suggests that displacement isn't uniform; it’s really about where the movement is concentrated at any given time.
Title and authors: Lu: And they also map these topological features back to specific neighbourhoods and years using TCripser, which gives them the spatio-temporal coordinates for the birth and death of each feature. This lets them pinpoint exactly what constitutes each connected component at a given filtration parameter value.
Jane: So, if I'm listening, I should be thinking about how these features map back to specific places and times, like how neighborhood "twenty-seven Atocha" pops up as a focal point from two thousand nine in the H0 features for Madrid <ref:2512.10753#pg2>. That shows they can find both large-scale and localized dynamics.
Tom: Exactly. It confirms that this method reliably identifies displacement in the Madrid case study by capturing patterns richer than just simple spatio-temporal proximity indicators, which is what this paper is really showing us.
Lu: The persistence of H1 and H2 features suggests how long it takes for a hole to close, indicating that neighbourhoods and years with those features present different characteristics from their neighbors, which drives different behaviors.
Jane: So what does this mean for the world outside of Madrid? The paper emphasizes that by using publicly available data collected in most countries, their approach is highly replicable and opens the door to applying this methodology to measure displacement in cities around the world.
Tom: It suggests we can use this topological approach not just for one city but as a general framework for quantifying population displacement as a major toll in social urban transformations.
Meng: I still have my operational concerns about scaling this up; how much compute does running these cubical complexes at scale cost, and what happens to the accuracy when we move to real-time traffic data instead of historical address changes?
Lu: The paper acknowledges that the construction of the T-construction for cubical complexes follows a top-down approach, and they detail those preprocessing steps to ensure accuracy regarding administrative changes. They also suggest future variations could involve splitting the 'city' group into those who move to an adjacent neighborhood and the rest of the city.
Jane: That's a good point about future work. It shows the researchers aren't just stopping at Madrid; they are looking at how they could refine the model, perhaps by adding socioeconomic features to enrich that persistent homology study.
Title and authors: Tom: I also want to mention that they did address some potential issues with neighborhood boundaries, detailing how data preprocessing handles them so no 'holes' are created in the map when dealing with newly created administrative units.
Lalam: I find the results on H2 features in the 'stay' group indicating significant population displacement since two thousand thirteen to be a concrete finding that moves beyond just describing what’s there; it quantifies a specific period of change based on these topological metrics <ref:2512.10753#pg2>.
Jane: So, wrapping up, this paper shows how to use complex topological analysis on address data to reliably identify and map specific patterns of population displacement across different groups and time scales in a way that's highly replicable.
Tom: It’s a solid methodology because it moves away from ad-hoc metrics and relies on the properties of persistent homology to extract spatial-temporal dynamics that are hard to see otherwise.
Lu: And as they suggest, exploring how persistent H1 and H2 features indicate the duration for which a hole remains open gives us insight into the specific characteristics that drive different behaviors in different neighbourhoods.
Meng: From an engineering standpoint, it’s a powerful way to detect subtle population dynamics, like a temporary easing during the COVID-nineteen pandemic in central areas <ref:2512.10753#pg1>. That kind of nuance is what we need when we deploy models to real traffic data.
Jane: So the main implication for us listening here is that this method opens up a way to quantify displacement as a major toll in social urban transformations using tools that are more grounded than just simple statistical proximity measures.
Tom: We're looking forward to seeing how researchers apply this approach to cities beyond Madrid, which is where the real test of its generalizability will be.
Lu: The paper opens the door to applying this same methodology to measure displacement in cities around the world, provided you handle the data preprocessing carefully as they do.
Meng: I just hope when we start building production systems for this, we can figure out a way to make that cubical complex construction more efficient so it doesn't take too much time.
Jane: It’s a lot of technical detail, but the core idea is powerful: using persistent homology on four cubical complexes simultaneously to quantify displacement patterns not observable from raw address change data.
The paper's summary: Tom: So, we’re looking at the paper "Quantifying displacement: an urban expansion consequence via persistent homology." It’s about taking raw address change data and using this topological method to actually measure how people are being pushed around in a city.
Jane: Right, so if you think about it, the main idea is that we can look at the movement of people over time and space, not just as a list of moves, but as some kind of shape or structure.
Lu: Exactly. They’re building these four cubical complexes simultaneously to capture the geography and the timeline all at once in a very structured way.
Meng: I’m interested in how they handle the data construction because that's where most of my problems start—the administrative boundaries shifting, right?
Tom: Yeah, and they actually spend a lot of time detailing that preprocessing step. They mention how they merge people moving within a district into all the neighborhoods in Vicálvaro from two thousand four to two thousand sixteen.
Jane: It seems like they’re trying to make the method as flexible as possible across different cities, which is what they claim by not relying on some super specific city metrics.
Lu: The core novelty here is using Persistent Homology to quantify this displacement. It’s a tool that lets you look at the structure of these complexes and see persistent features that aren't obvious in the raw data.
Tom: So, they translate those topological features into something measurable, like H0, H1, and H2 holes. These holes tell us about different kinds of movement dynamics.
Jane: And what they found is that these features actually point to specific neighborhoods and years that are behaving differently than the rest of the city.
Meng: If I’m running this at scale, I need to know if these features are stable or if they just fluctuate wildly because you’re dealing with noisy data from real-world traffic instead of clean administrative records.
Tom: Well, they do have some caveats about the time it takes to build these complexes depending on the country and the level of detail in their input data, which is a fair point for deployment.
Lu: But looking at their results, they found that for instance, neighborhood "twenty-seven Atocha" keeps showing up as a focal point across all four movement groups over those twenty years.
Jane: That’s the tangible result there—a specific place that acts like an anchor point for displacement patterns.
Tom: And it shows how those higher dimensional features, like H2 features in the 'stay' group, are linked to significant population shifts in the central city area since two thousand thirteen.
Meng: So what does this mean for someone just listening? It means that when we look at a city’s history, we can find these specific hotspots of pressure rather than just a general statistic about migration flow.
Jane: Precisely. It moves the conversation from "people moved" to "this specific neighborhood experienced this kind of structural change during this time."
Lu: And they use tools like TCripser to map those abstract topological features back onto real coordinates, giving us the exact x, y, z location and year for each feature birth and death.
Tom: That mapping is crucial because it validates the abstract math with a physical location we can actually point to on a map.
Jane: And ultimately, this approach suggests that displacement isn't just a number; it’s a complex pattern embedded in the shape of how people live and move within those urban structures.
Meng: I still see the challenge in making sure this method is efficient enough for real-time monitoring, but getting these localized dynamics mapped is definitely something worth exploring further.
The paper's improvements: Tom: So, we’re talking about the improvements suggested in this paper for their displacement quantification method. It sounds like they aren't just stopping at what they did with Madrid data, right?
Jane: Right, so the paper points out that their methodology isn't perfect as is and suggests a few ways to make it better or more flexible.
Lu: They suggest splitting the 'city' group further into two parts: those who move to an adjacent neighborhood and then the rest of the city. That would give them finer resolution on internal city movement.
Meng: I like that idea because it addresses a potential weakness in their current grouping—they treat all 'city' movers as one homogenous blob, which I know breaks down when you look at actual deployment scenarios.
Tom: And they also suggest exploring socioeconomic features of the displaced populations to enrich the persistent homology study. Adding that kind of data would give them a much deeper view of *why* people are moving, not just where they are going.
Jane: So, it’s about adding more context to the topological shape they’re measuring; not just counting holes but understanding the social reasons behind those holes forming or closing.
Lu: That would definitely increase the complexity of the model training, but it would make the interpretation much richer, allowing them to distinguish between different types of displacement pressure.
Tom: From an engineering standpoint, I see that adding these features means we need a more complex representation learning step for the AI to process everything at once.
Meng: And that complexity is what I worry about when we try to deploy something; how much extra compute does that add to the inference time? We need efficiency, not just richness.
Jane: The authors acknowledge those limitations, saying that while they can capture subtle patterns like temporary easing during COVID-nineteen the method still needs refinement for broader application.
Lu: They also talk about how data preprocessing might vary by country and how that impacts the time it takes to construct those cubical complexes. That’s a practical constraint they are trying to manage.
Tom: So, the improvement isn't just a new equation; it’s a whole strategy for making this topological tool work reliably across different urban environments globally.
Jane: It suggests that the future work will focus on making the input data handling more robust so that when we apply this framework to other cities, we don't run into those boundary issues they mentioned.
Meng: I'm hoping they explore how these features interact with real-time data streams, because right now, running this analysis over historical address changes is fine for a study but not for monitoring a live housing market pressure spike.
Lu: That would be the next logical step—moving from retrospective quantification to prospective monitoring of housing tension.
Conclusion: Tom: So, to wrap up, this paper "Quantifying displacement: an urban expansion consequence via persistent homology" shows how we can move beyond simple data points to map out actual patterns of involuntary residential dislocation in cities.
Jane: Exactly. It’s about using complex topology to find those specific neighborhoods and time periods that are under pressure, which is a much more concrete way to understand urban change than just looking at raw address changes.
Lu: The core idea is that by constructing these cubical complexes and using persistent homology, we can quantify displacement in a way that’s not dependent on some arbitrary city-specific metric.
Meng: I think the most important thing is seeing how this method handles the complexity of real-world administrative changes, because if it breaks there at scale, it’s useless for production.
Lalam: I'm very sure that by using these topological features, we are capturing displacement dynamics that were simply not visible in standard statistical proximity measures before.
Tom: And the result is a clear pattern: neighborhood "twenty-seven Atocha" acting differently throughout the twenty years studied, which gives us a specific case study to look at.
Jane: So what does this mean for someone just listening? It means that urban studies can get much more detailed about *where* and *when* people are actually being displaced in a city.
Lu: And the future work suggested is pretty clear: they need to explore adding socioeconomic features to these topological studies to see how those external factors influence the spatial dynamics.
Meng: I’m still thinking about the cost of running this analysis; if we want this in production, we need an efficient way to calculate those persistent homologies without it taking too long.
Tom: Yeah, that’s a big engineering hurdle, but the foundation they laid here is solid for applying this kind of structural analysis to other cities worldwide.
Jane: This study really opens the door for applying this topological approach to measure displacement in many different urban contexts beyond Madrid.
Lu: And it confirms that persistent H1 and H2 features tell us about the duration a specific structural characteristic persists, which drives different behaviors in those neighborhoods.
Lalam: I'm confident that by quantifying these spatial-temporal structures, we are getting a much more accurate picture of the human toll of urban expansion.
Tom: So, this paper demonstrates how to use persistent homology on four cubical complexes simultaneously to quantify displacement patterns not observable from raw address change data.
Jane: It’s a lot of technical detail, but the main implication is that we have a new way to quantify displacement as a major toll in social urban transformations.
Lu: We’re definitely looking forward to seeing how researchers apply this methodology to cities beyond Madrid, testing its generalizability across different data sets.
Meng: I just hope when we start building production systems for this, we can figure out a way to make that cubical complex construction more efficient so it doesn't take too much time.
Tom: It’s a lot of technical detail, but the core idea is powerful: using persistent homology on four cubical complexes simultaneously to quantify displacement patterns not observable from raw address change data.
More episodes
- 2505.16254-Human-AI Writing as Distributed Cognition: Reassessing Collaborative Writing Theories in the Age of LLMs
- 2506.10913-Choreographic Quick Changes: First-Class Location (Set) Polymorphism
- 2507.14812-A Black-Box Approach for Exogenous Replenishment in Online Resource Allocation
- 2507.15448-Frames over finite fields with respect to the Galois inner product and their connections to self-orthogonal and LCD codes
- 2508.21572-NewsReX: An Open-Source Multi-Framework for Neural News Recommendation
- 2509.17028-Impact of Packetization on Network Calculus Analysis
- 2510.17342-AoA Services in 5G Networks: A Framework for Real-World Implementation and Systematic Testing
- 2511.04088-Efficient and rate-optimal list-decoding in the presence of minimal feedback
- 2511.17264-Notes on Stack Machines and Quantum Stack Machines
- 2601.15318-On the closest balanced game