Location prediction in social media based on tie strength

McGee, Jeffrey; Caverlee, James; Cheng, Zhiyuan

doi:10.1145/2505515.2505544

Cited by 126 publications

(107 citation statements)

References 17 publications

Supporting

Mentioning

102

Contrasting

Order By: Relevance

“…Gazetteers and Geographical Databases are also well applied to the study and some tools used include the United States board on geographic names popularly called GeoNames 2 , GeoNet 3 and the US census TIGER Gazetteers 4 . Some works have also used a hybrid of the earlier mentioned techniques.…”

Section: Gazetteersmentioning

confidence: 99%

“…Corpus Size Period Covered Duration (Months) [14] 2,495,000 Jan '11 -May '11 5 [32] 380,000 Mar '10 1 [5] 62,000 Apr '10 -May '10 2 [15] 47,700,000 Apr '12 -Nov '12 8 [16] 4,330,000 Jun '10 1 [33] 1,524,000 Jul '11 -Aug '11 2 [4] 100,000,000 Jun '10 1 [23] 20,000,000 Apr '11 1 [31] 615,000,000 Jun '10 -Apr '11 13 [17] 80,000,000 Sep '11 -Feb '12 6 …”

Section: Referencementioning

confidence: 99%

See 1 more Smart Citation

A survey of location inference techniques on Twitter

Ajao

Hong

Liu

2015

Journal of Information Science

132

View full text Add to dashboard Cite

The increasing popularity of the social networking service, Twitter, has made it more involved in day-to-day communications, strengthening social relationships and information dissemination. Conversations on Twitter are now being explored as indicators within early warning systems to alert of imminent natural disasters such earthquakes and aid prompt emergency responses to crime. Producers are privileged to have limitless access to market perception from consumer comments on social media and microblogs. Targeted advertising can be made more effective based on user profile information such as demography, interests and location. While these applications have proven beneficial, the ability to effectively infer the location of Twitter users has even more immense value. However, accurately identifying where a message originated from or author's location remains a challenge thus essentially driving research in that regard. In this paper, we survey a range of techniques applied to infer the location of Twitter users from inception to state-of-the-art. We find significant improvements over time in the granularity levels and better accuracy with results driven by refinements to algorithms and inclusion of more spatial features.

show abstract

Section: Gazetteersmentioning

confidence: 99%

Section: Referencementioning

confidence: 99%

A survey of location inference techniques on Twitter

Ajao

Hong

Liu

2015

Journal of Information Science

132

View full text Add to dashboard Cite

show abstract

“…Rout et al [22] takes into account the populations of locations. McGee et al [17] classified informative friends that live close to the target user and others. Yamaguchi et al [27] utilized users who attract a lot of local attentions for location inference.…”

Section: Related Workmentioning

confidence: 99%

Online User Location Inference Exploiting Spatiotemporal Correlations in Social Streams

Yamaguchi

Amagasa

Kitagawa

et al. 2014

Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management

View full text Add to dashboard Cite

The location profiles of social media users are valuable for various applications, such as marketing and real-world analysis. As most users do not disclose their home locations, the problem of inferring home locations has been well studied in recent years. In fact, most existing methods perform batch inference using static (i.e., pre-stored) social media contents. However, social media contents are generated and delivered in real-time as social streams. In this situation, it is important to continuously update current inference results based on the newly arriving contents to improve the results over time. Moreover, it is effective for location inference to use the spatiotemporal correlation between contents and locations. The main idea of this paper is that we can infer the locations of users who simultaneously post about a local event (e.g., earthquakes). Hence, in this paper, we propose an online location inference method over social streams that exploits the spatiotemporal correlation, achieving 1) continuous updates with low computational and storage costs, and 2) better inference accuracy than that of existing methods. The experimental results using a Twitter dataset show that our method reduces the inference error to less than 68% of existing methods. The results also show that the proposed method can update inference results in constant time regardless of the amount of accumulated contents.(a) In ordinary times.(b) After the earthquake at Hiroshima. Figure 1: Geographical distributions of tweets containing the word "earthquake" (a) in ordinary times, and (b) after the earthquake at Hiroshima prefecture. Each circle represents the ratio of the number of tweets posted there. The distribution of the word occurrence changes over time.

show abstract

“…Some recent studies have addressed this issue by introducing social relationship strength modeling. Recent relationship strength models have been using interaction activity [20], profile similarity [23], amount of shared time [6], geographical location [14], and mutual friends [11]. However, we believe that to build multimedia-based stories we need a new definition of relationship strength that includes photos metadata.…”

Section: Related Workmentioning

confidence: 99%

Towards Storytelling by Extracting Social Information from OSN Photo's Metadata

Saini

Alzamzami

Saddik

2014

Proceedings of the First International Workshop on Internet-Scale Multimedia Management

View full text Add to dashboard Cite

The popularity of online social networks (OSNs) is growing rapidly over time. People share their experiences with their friends and relatives with the help of multimedia such as image, video, text, etc. The amount of such shared multimedia is also growing likewise. The large amount of multimedia data on OSNs contains in it a snapshot of user's life. This social network data can be crawled to build stories about individuals. However, the information needed for a story, such as events and pictures, is not fully available on user's own profile. While part of this information can be retrieved from user's own timeline, a large amount of event and multimedia information is only available on friend's profiles. As the number of friends can be very large, in this work we focus on identifying subset of friends for enriching the story data. In this paper we explore social relationships from multimedia perspective and propose a framework to build stories using information from multiple-profiles. To the best of our knowledge, this is the first work on building stories using multiple OSN profiles. The experimental results show that with the proposed method we get more information (events, locations, and photos) about the individuals in comparison to the traditional methods that rely on user's own profile alone.

show abstract

Location prediction in social media based on tie strength

Cited by 126 publications

References 17 publications

A survey of location inference techniques on Twitter

A survey of location inference techniques on Twitter

Online User Location Inference Exploiting Spatiotemporal Correlations in Social Streams

Towards Storytelling by Extracting Social Information from OSN Photo's Metadata

Contact Info

Product

Resources

About