GSAM: A deep neural network model for extracting computational representations of Chinese addresses fused with geospatial feature

Automatic Identification of Addresses: A Systematic Literature Review

“…Computation linguistics, attention mechanisms, LSTMs, and location-based services emerge as some of the most recent research topics. As shown in Table 4, four clusters were identified, based on VOSviewer default clustering technique [17]: address matching and NLP, in green (e.g.,: Xu et al [20]), GIS/geocoding and machine learning, in blue (e.g.,: Peng et al [21]), address standardization, in yellow (e.g.,: Churches et al [6]) and address recognition and parsing, in red (e.g.,: Wei et al [22]). VOSviewer was also used to perform the analysis on co-authorship.…”

“…Within the more recently published papers considered in the present literature review, the most relevant opportunities for further work can be summarized as follows: the use of representative and large enough datasets [20]; the inclusion of duplicate place names, in order to enable the application of the proposed methodology to a national address database [9]; to improve accuracy, different weights might be assigned to the addresselement vectors depending on their hierarchy [9]; the need to fine tune the weight ratio of fused features, such as coordinates and the semantic representation of addresses, alongside the improvement of the underlying concatenation method and measurement metrics [20]; the adoption of systematic approaches for tuning hyper-parameters and experimenting with different architectures [35]; the need to involve more complex spatial objects and relations [2,8]. Some of the limitations highlighted in less recent studies, however, should also be taken in consideration in the application of the most recent methods, like the need to tackle privacy and confidentiality issues [51] when using personal quasi-identifiers such as addresses (especially, residential ones).…”

“…In two of the most recent studies included in the present literature review also based on deep learning methods, bidirectional encoder representations from Transformers (BERT) are proposed instead. The first one is the study by Xu et al [20], which proposes a method for fusing addresses and geospatial data based on BERT (in what concerns the learning of addresses' semantics) and a K-Means high-dimensional clustering algorithm, enhanced by innovative fine-tuning techniques, to create a geospatial-semantic address model (GSAM). The computational representation extracted from GSAM is then employed for predicting address location, based on a neural network architecture to minimize the Euclidean distance between the predicted and real coordinates.…”

“…In terms of performance, the GSAM model [20] achieves a classification accuracy above 0.97, against a minimum expected accuracy of 0.91 by other methods; the BERT-CRF model [37] achieves the highest F1 score on generalization ability (0.78), when compared to benchmark models combining word2vec, BiLSTM, and CRF methods (with an average F1 score of 0.41), as well as an equally high F1 score on the testing dataset (0.95).…”

A GloVe-Based POI Type Embedding Model for Extracting and Identifying Urban Functional Regions

“…The aforementioned works have focused on examining the spatial features of cities at a micro level. However, as urban functional regions are the result of the interaction between humans and the land surface and the spatial accumulation of social and economic features in cities [3,30], the features of a single place type cannot completely describe the functional types of an urban region. Yao et al [15] modeled the relationship between spatial distributions of POIs and land use types based on word2vec.…”