On-line aggregation of POIs from Google and Facebook

Toccu, Maurizio; Psaila, Giuseppe; Altomare, Davide

doi:10.1145/3297280.3297576

Cited by 2 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main focus of the research is to find the proper formulation for the MatchingPlaces fuzzy relation, that evaluates if two place descriptors actually describe the same place. In [9], we introduced a formulation, that we report in the following definition. This formulation, hereafter denoted as MatchingPlaces V1 , constitutes our baseline.…”

Section: Baseline Formulation For the Matchingplaces Relationmentioning

confidence: 99%

“…In [9], we tested the effectiveness of this relation with a threshold α = 0.75 and with thresholds mind = 100 m and maxd = 2000 m for the Close relation. We obtained good results, but the data set used during the experiments was clean: no descriptors with missing fields were present.…”

Section: Definition 6 Relation Matchingplacesmentioning

confidence: 99%

“…In order to evaluate the effectiveness of our technique, we built three test data sets-specifically, by exploiting the algorithm to download place descriptors proposed in [9]. We downloaded place descriptors from Google Places; for each of them, we looked for up to two descriptors of Facebook pages, possibly corresponding to the place described by the Google Places descriptor.…”

Section: Experimental Evaluationmentioning

confidence: 99%

“…Table 5 reports, in the upper part, the results obtained by applying the MatchingPlaces V1 relation (see Definition 6), that is the baseline definition of the MatchingPlaces relation; in the middle, the table reports the results obtained for the MatchingPlaces V2 relation with Con f 14 as reference configuration. Recall (from Section 5.2) that, as far as the Close relation is concerned, we run experiments for the MatchingPlaces V1 relation with settings mind = 100 m and maxd = 2000 m, as in [9]; furthermore, the minimum threshold to defuzzify is α = 0.75. Table 5.…”

Section: Con F β Geo αmentioning

confidence: 99%

“…Then, we downloaded three data sets describing places in three different cities, such as Manchester (UK), Genoa (Italy), and Stuttgart (Germany): descriptors were downloaded from Google Places and paired with descriptors of Facebook pages. As a baseline for the evaluation, we considered our preliminary version of the binary fuzzy relation introduced in [9], showing that the new definition (proposed in this paper) significantly improves the effectiveness of the technique. Then, we make a comparison with a machine-learning technique, named "random forest": it is a well known classification technique, to be executed off-line.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Fuzzy Technique for On-Line Aggregation of POIs from Social Media: Definition and Comparison with Off-Line Random-Forest Classifiers

Psaila

Toccu

2019

Information

Self Cite

View full text Add to dashboard Cite

Social media represent an inexhaustible source of information concerning public places (also called points of interest (POIs)), provided by users. Several social media own and publish huge and independently-built corpora of data about public places which are not linked each other. An aggregated view of information concerning the same public place could be extremely useful, but social media are not immutable sources, thus the off-line approach adopted in all previous research works cannot provide up-to-date information in real time. In this work, we address the problem of on-line aggregating geo-located descriptors of public places provided by social media. The on-line approach makes impossible to adopt machine-learning (classification) techniques, trained on previously gathered data sets. We overcome the problem by adopting an approach based on fuzzy logic: we define a binary fuzzy relation, whose on-line evaluation allows for deciding if two public-place descriptors coming from different social media actually describe the same public place. We tested our technique on three data sets, describing public places in Manchester (UK), Genoa (Italy) and Stuttgart (Germany); the comparison with the off-line classification technique called “random forest” proved that our on-line technique obtains comparable results.

show abstract

Section: Baseline Formulation For the Matchingplaces Relationmentioning

confidence: 99%

Section: Definition 6 Relation Matchingplacesmentioning

confidence: 99%

Section: Experimental Evaluationmentioning

confidence: 99%

Section: Con F β Geo αmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations