A Deep Learning Model of Spatial Distance and Named Entity Recognition (SD-NER) for Flood Mark Text Classification

Aug 2, 2026

Upload by Finn

Information on historical flood levels is communicated verbally, in documents, or through flood marks, which are especially valuable for public awareness and quantitative flood modeling. Flood-mark information is increasingly available on the Internet but is difficult to locate and classify. The work proposes an ensemble deep learning SD-NER model that combines named entity recognition, deep neural networks, and spatial analysis using a matrix of minimum distances between toponyms. Tested on Poland, the model reaches an F1 score of 0.920, improving performance by 17% versus models without the spatial module.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Introduction

Flood marks and their value

Challenges of finding Internet sources

Proposed SD-NER approach

water Article A Deep Learning Model of Spatial Distance and Named Entity Recognition (SD-NER) for Flood Mark Text Classiﬁcation Robert Szczepanek Citation: Szczepanek, R. A Deep Learning Model of Spatial Distance and Named Entity Recognition (SD-NER) for Flood Mark Text Classiﬁcation. Water 2023, 15, 1197 . [https://doi.org/10.3390/w15061197](https://doi.org/10.3390/w15061197) Academic Editor: Jianjun Ni Received: 8 February 2023 Revised: 12 March 2023 Accepted: 16 March 2023 Published: 20 March 2023 Copyright: © 2023 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . Institute of Geological Sciences, Faculty of Geography and Geology, Jagiellonian University, 30-387 Krakow, Poland; [robert.szczepanek@uj.edu.pl](robert.szczepanek@uj.edu.pl) Abstract: Information on historical ﬂood levels can be communicated verbally, in documents, or in the form of ﬂood marks. The latter are the most useful from the point of view of public awareness building and mathematical modeling of ﬂoods. Information about ﬂood marks can be found in documents, but nowadays, they are starting to appear more often on the Internet. The only problem is ﬁnding them. The aim of the presented work is to create a new model for classifying Internet sources using advanced text analysis (including named entity recognition), deep neural networks, and spatial analysis. As a novelty in models of this type, it was proposed to use a matrix of minimum distances between toponyms (rivers and towns/villages) found in the text. The resulting distance matrix for Poland was published as open data. Each of the methods used is well known, but so far, no one has combined them into one ensemble machine learning model in such a way. The proposed SD-NER model achieved an F1 score of 0.920 for the binary classiﬁcation task, improving the model without this spatial module by 17% . The proposed model can be successfully implemented after minor modiﬁcations for other classiﬁcation tasks where spatial information about toponyms is important. Keywords: machine learning; ensemble model; cultural heritage; ﬂood memory; Poland; open data; convolutional neural networks; natural language processing; high-water ﬂood marks 1. Introduction Searching for information on the Internet is not a simple task. The results returned by popular search engines for ambiguous phrases often contain unexpected results. In January 2023, out of the ﬁrst 10 search results for the phrase “ﬂood mark”, only 5 in Google Search and 8 in DuckDuckGo refer to ﬂood issues. Search results depend on the search engine used, and most importantly, they change over time [1] . With dozens or even hundreds of thousands of websites returned by search engines, it is extremely difﬁcult to determine which of them contain information of interest to us. The task that each user faces is to classify the search results into results that meet their expectations and the others. Most often we do it intuitively based on previous experiences. For example, we ignore results that are described as advertising and those that meet the search criteria, but are not of interest to us. Fortunately, modern machine learning algorithms based on natural language processing (NLP) make this classiﬁcation much easier [2–4], no matter what information we are looking for. Flood marks (or high-water marks) are permanent graphic information describing and often also showing how high water reached during catastrophic ﬂoods [5,6] . Sometimes they are the only evidence of extreme events because no written records have survived. Quantitative information about ﬂood events, especially those from many years ago, is often forgotten. This applies to both print

FAQs

What problem does the SD-NER model address in flood mark research?

It targets the difficulty of finding and classifying flood-mark information on the Internet, where relevant sources are scattered and search results are noisy.

What is novel about the proposed SD-NER model?

It introduces a spatial module that uses a matrix of minimum distances between toponyms (such as rivers and towns/villages) extracted from the text.

How well does the model perform for binary classification?

It achieves an F1 score of 0.920 for binary classification, improving results by 17% compared with a model without the spatial distance module.

Aug 2, 2026

Upload by Finn