Ábel Elekes

Mr. Ábel Elekes

None

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
8 March 2021 in Sustainability
Verified
Journal Article
Open access
PDF file
Zoltán KovácsZoltán Kovács
Zoltán Kovács
GVGyörgy Vida
György Vida
Ábel ElekesÁbel Elekes
Ábel Elekes
Tamás KovalcsikTamás Kovalcsik
Tamás Kovalcsik
Despite the growing importance of mobile tracking technology in urban planning and traffic forecasting, its utilization in the understanding of the basic laws governing tourist flows remains limited. Knowledge regarding the motivations and spatial behavior of tourists has great potential in sustainable tourism studies. In this paper, we combine social media (Twitter) and mobile positioning data (MPD) in the analysis of international tourism flows in Szeged, a secondary urban center in Hungary. First, the content of geotagged and non-geotagged Twitter messages referring to Szeged in a six-month period of 2018 was analyzed. In this way specific events attracting foreign tourists were identified. Then, using MPD data of foreign SIM cards, visitor peaks in the investigated period were defined. With the joint application of the social media and mobile positioning analytical tools, we were able to identify those attractions (festivals, sport and cultural events, etc.) that generated significant tourism arrivals in the city. Furthermore, using the mixed-method approach we were also able to analyze the movements of foreign visitors during one large-scale tourism event and evaluate its hinterland. Overall, this study supports the idea that social media data should be combined with other real-time data sources, such as MPD, in order to gain a more precise understanding of the behavior of tourists. The proposed analytical tool can contribute to methodological and conceptual development in the field, and information gained by its application can positively influence not only tourism management and planning but also tourism marketing and placemaking.

1 June 2017
Verified
Conference Paper
Open access
Ábel ElekesÁbel Elekes
Ábel Elekes
MSMartin Schaeler
Martin Schaeler
KBKlemens Boehm
Klemens Boehm
Finding similar words with the help of word embedding models, such as Google's Word2Vec or Glove, computed on large-scale digital libraries has yielded meaningful results in many cases. However, the underlying notion of similarity has remained ambiguous. In this paper, we examine when exactly similarity values in word embedding models are meaningful. To do so, we analyze the statistical distribution of similarity values systematically, conducting two series of experiments. The first one examines how the distribution of similarity values depends on the different embedding-model algorithms and parameters. The second one starts by showing that intuitive similarity thresholds do not exist. We then propose a method stating which similarity values actually are meaningful for a given embedding model. In more abstract terms, our insights give way to a better understanding of the notion of similarity in embedding models and to more reliable evaluations of such models.

25 October 2022 in Sustainability
Verified
Journal Article
Open access
PDF file
Tamás KovalcsikTamás Kovalcsik
Tamás Kovalcsik
Ábel ElekesÁbel Elekes
Ábel Elekes
Lajos BorosLajos Boros
Lajos Boros
LKLászló Könnyid
László Könnyid
Zoltán KovácsZoltán Kovács
Zoltán Kovács
Even though tourism (both domestic and international) is one of the main triggering factors of human mobility worldwide, some of its forms are unexplored. This can be partly linked with the lack of reliable data and obstacles related to data processing and interpretation. Mobile Positioning Data (MPD) allows us to identify various forms of tourism that are undetectable through traditional data sources such as accommodation statistics. Using MPD, not only same-day tourists but also the real time mobility patterns of tourists among various destinations can be revealed, and even hidden (i.e., unobserved) forms of tourism can be detected. However, despite the obvious benefits of such data, very few comprehensive studies exist to date on the processing, and interpretation of MPD in tourism research. In this paper, a case study is presented on the challenges and opportunities of processing MPD from raw to good quality researchable data offering a baseline tool for MPD-based research in the field of tourism. With the methodology introduced in this paper, it is possible to provide a more accurate picture of tourist flows regarding unobserved tourists, including same-day visitors.

1 June 2019
Verified
Conference Paper
Closed access
Ábel ElekesÁbel Elekes
Ábel Elekes
ASAntonino Simone Di Stefano
Antonino Simone Di Stefano
MSMartin Schaler
Martin Schaler
KBKlemens Bohm
Klemens Bohm
MKMatthias Keller
Matthias Keller
Text classification helps to categorize large number of documents in Digital Libraries. Text classification results highly depend on the quality of labeled training data. In practice, the process of manually annotating documents is a hidden cost that is often overlooked. We propose a general preprocessing method for scenarios in which training data is scarce. It clusters semantically similar terms by including both a semantic distance measure and a probabilistic model of any task-specific term distributions. By preprocessing the training data with our method, one increases the mean classification performance of all tested classification approaches in text classification tasks having 500 or 1000 training samples. The largest observed increase is 15%. When more training samples are available, we report significant improvements in most scenarios as well.

1 January 2018
Verified
Conference Paper
Closed access
Ábel ElekesÁbel Elekes
Ábel Elekes
AEAdrian Englhardt
Adrian Englhardt
MSMartin Schäler
Martin Schäler
KBKlemens Böhm
Klemens Böhm
Word embeddings are powerful tools that facilitate better analysis of natural language. However, their quality highly depends on the resource used for training. There are various approaches relying on n-gram corpora, such as the Google n-gram corpus. However, n-gram corpora only offer a small window into the full text – 5 words for the Google corpus at best. This gives way to the concern whether the extracted word semantics are of high quality. In this paper, we address this concern with two contributions. First, we provide a resource containing 120 word-embedding models – one of the largest collection of embedding models. Furthermore, the resource contains the n-gramed versions of all used corpora, as well as our scripts used for corpus generation, model generation and evaluation. Second, we define a set of meaningful experiments allowing to evaluate the aforementioned quality differences. We conduct these experiments using our resource to show its usage and significance. The evaluation results confirm that one generally can expect high quality for n-grams with n > 3.