Twitter alloy steel disambiguation and user relevance via one-class and two-class news titles classifiers

Abstract

This paper addresses the nontrivial task of Twitter financial disambiguation (TFD), which is relevant to filter financial domain tweets (e.g., alloy steel or coffee prices) when no unique identifiers (e.g., cashtags) are adopted. To automate TFD, we propose a transfer learning approach that uses freely labeled news titles to train diverse one-class and two-class classification methods. These include different text handling transforms, adaptations of statistical measures and modern machine learning methods, including support vector machines (SVM), deep autoencoders and multilayer perceptrons. As a case study, we analyzed the domain of alloy steel prices, collecting a recent Twitter dataset. Overall, the best results were achieved by a two-class SVM fed with TFD statistical measures and topic model features, obtaining an 80% and 71% discrimination level when tested with 11,081 and 3000 manually labeled tweets. The best one-class performance (78% and 69% for the same test tweets) was obtained by a term frequency-inverse document frequency classifier (TF-IDFC). These models were further used to generate a Financial User Relevance rank (FUR) score, aiming to filter relevant users. The SVM and TF-IDFC FUR models obtained a predictive user discrimination level of 80% and 75% when tested with a manually labeled test sample of 418 users. These results confirm the proposed joint TFD-FUR approach as a valuable tool for the selection of Twitter texts and users for financial social media analytics (e.g., sentiment analysis, detection of influential users).

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Notes

  1. 1.

    https://blog.hootsuite.com/twitter-statistics/.

  2. 2.

    https://www.focus-economics.com/blog/steel-facts-commodity-explainer.

  3. 3.

    https://money.cnn.com/2018/03/07/news/companies/trump-tariffs-steel-jobs/index.html.

  4. 4.

    https://kallanish.com/en/.

  5. 5.

    https://www.steelorbis.com/.

  6. 6.

    https://www.nytimes.com/.

  7. 7.

    https://www.reuters.com/.

  8. 8.

    https://www.kaggle.com/therohk/million-headlines/home.

  9. 9.

    https://github.com/paolazola/Twitter-Financial-Disambiguation-Financial-Users-Relevance.

References

  1. 1.

    Awwalu J, Bakar AA, Yaakub MR (2019) Hybrid n-gram model using naïve bayes for classification of political sentiments on twitter. Neural Comput Appl 1:1–14

    Google Scholar 

  2. 2.

    Zola P, Cortez P, Carpita M (2019) Twitter user geolocation using web country noun searches. Decis Support Syst 120:50–59

    Article  Google Scholar 

  3. 3.

    Oliveira N, Cortez P, Areal N (2017) The impact of microblogging data for stock market prediction: using twitter to predict returns, volatility, trading volume and survey sentiment indices. Expert Syst Appl 73:125–144

    Article  Google Scholar 

  4. 4.

    Groß-Klußmann A, König S, Ebner M (2019) Buzzwords build momentum: global financial twitter sentiment and the aggregate stock market. Expert Syst Appl 136(1):171–186

    Article  Google Scholar 

  5. 5.

    Pagolu VS, Reddy KN, Panda G, Majhi B (2016) Sentiment analysis of twitter data for predicting stock market movements. In: International conference on signal processing, communication, power and embedded system (SCOPES). IEEE, pp 1345–1350

  6. 6.

    Lechthaler F, Leinert L (2012) Moody oil: What is driving the crude oil price? Empirical Economics 1:1–32

    Google Scholar 

  7. 7.

    Li J, Xu Z, Yu L, Tang L (2016) Forecasting oil price trends with sentiment of online news articles. Procedia Comput Sci 91:1081–1087

    Article  Google Scholar 

  8. 8.

    Bollen J, Mao H, Zeng X (2011) Twitter mood predicts the stock market. J Comput Sci 2(1):1–8

    Article  Google Scholar 

  9. 9.

    Feuerriegel S, Neumann D (2013) News or noise? how news drives commodity prices. In: Proceedings of the international conference on information systems, ICIS, Milano, Italy, December 15–18

  10. 10.

    Rao T, Srivastava S (2013) Modeling movements in oil, gold, forex and market indices using search volume index and twitter sentiments. In: Proceedings of the 5th annual ACM web science conference. ACM, pp 336–345

  11. 11.

    Pröllochs N, Feuerriegel S, Neumann D (2015) Enhancing sentiment analysis of financial news by detecting negation scopes. In: 48th Hawaii international conference on system sciences. IEEE, pp 959–968

  12. 12.

    Nguyen TH, Shirai K, Velcin J (2015) Sentiment analysis on social media for stock movement prediction. Expert Syst Appl 42(24):9603–9611

    Article  Google Scholar 

  13. 13.

    Daniel M, Neves RF, Horta N (2017) Company event popularity for financial markets using twitter and sentiment analysis. Expert Syst Appl 71:111–124

    Article  Google Scholar 

  14. 14.

    Maslyuk-Escobedo S, Rotaru K, Dokumentov A (2017) News sentiment and jumps in energy spot and futures markets. Pac-Basin Financ J 45:186–210

    Article  Google Scholar 

  15. 15.

    Huang D, Lehkonen H, Pukthuanthong K, Zhou G (2018) Sentiment across asset markets. SSRN 3185140. https://doi.org/10.2139/ssrn.3185140

  16. 16.

    Mudinas A, Zhang D, Levene M (2019) Market trend prediction using sentiment analysis: lessons learned and paths forward. CoRR arXiv:abs/1903.05440

  17. 17.

    Banerjee S, Pedersen T (2002) An adapted lesk algorithm for word sense disambiguation using wordnet. In: International conference on intelligent text processing and computational linguistics. Springer, pp 136–145

  18. 18.

    Zola P, Carpita M (2016) Forecasting the steel product prices with the arima model. Statistica and Applicazioni 14(1):1

    Google Scholar 

  19. 19.

    Wei W, Xia X, Wozniak M, Fan X, Damaševičius R, Li Y (2019) Multi-sink distributed power control algorithm for cyber-physical-systems in coal mine tunnels. Comput Netw 161:210–219

    Article  Google Scholar 

  20. 20.

    Lee C, Won J, Lee E-B (2019) Method for predicting raw material prices for product production over long periods. J Constr Eng Manag 145(1):05018017

    Article  Google Scholar 

  21. 21.

    Wei W, Song H, Li W, Shen P, Vasilakos A (2017) Gradient-driven parking navigation using a continuous information potential field based on wireless sensor network. Inf Sci 408:100–114

    Article  Google Scholar 

  22. 22.

    Pan S, Yang Q (2010) A survey on transfer learning. IEEE Trans Knowl Data Eng 22(10):1345–1359

    Article  Google Scholar 

  23. 23.

    Liu X, Zhou Y, Zheng R (2007) Sentence similarity based on dynamic time warping. In: Proceedings of the 1st IEEE international conference on semantic computing (ICSC), Irvine, California, USA, pp 250–256

  24. 24.

    Yan X, Guo J, Lan Y, Cheng X (2013) A biterm topic model for short texts. In: Proceedings of the 22nd international conference on World Wide Web, ACM, pp 1445–1456

  25. 25.

    Iosif E, Potamianos A (2015) Similarity computation using semantic networks created from web-harvested data. Nat Lang Eng 21(1):49–79

    Article  Google Scholar 

  26. 26.

    Kenter T, De Rijke M (2015) Short text similarity with word embeddings. In: Proceedings of the 24th ACM international on conference on information and knowledge management, ACM, pp 1411–1420

  27. 27.

    Song Y, Roth D (2015) Unsupervised sparse vector densification for short text similarity. In: Proceedings of the 2015 conference of the North American chapter of the association for computational linguistics: human language technologies, pp 1275–1280

  28. 28.

    Lee MD, Pincombe B, Welsh M (2005) An empirical evaluation of models of text document similarity. In: Proceedings of the annual meeting of the cognitive science society, pp 1254–1259

  29. 29.

    Chang M-W, Ratinov L-A, Roth D, Srikumar V (2008) Importance of semantic representation: dataless classification. AAAI 2:830–835

    Google Scholar 

  30. 30.

    Zhang H, Yang K, Jacob E (2015) Topic level disambiguation for weak queries. CoRR arXiv:abs/1502.04823

  31. 31.

    Amiri H, Resnik P, Boyd-Graber J, Daumé III H (2016) Learning text pair similarity with context-sensitive autoencoders. In: Proceedings of the 54th annual meeting of the association for computational linguistics (volume 1: Long Papers), vol 1, pp 1882–1892

  32. 32.

    Neculoiu P, Versteegh M, Rotaru M (2016) Learning text similarity with SIAMESE recurrent networks. In: Proceedings of the 1st workshop on representation learning for NLP, pp 148–157

  33. 33.

    Lim KH, Karunasekera S, Harwood A (2017) Clustop: A clustering-based topic modelling algorithm for twitter using word networks. In: IEEE international conference on big data (big data). IEEE, pp. 2009–2018

  34. 34.

    Chaplot DS, Salakhutdinov R (2018) Knowledge-based word sense disambiguation using topic models. In: Proceedings of the 32nd AAAI conference on artificial intelligence. (AAAI-18), pp 5062–5069

  35. 35.

    Li X, Zhang A, Li C, Ouyang J, Cai Y (2018) Exploring coherent topics by topic modeling with term weighting. Inf Process Manag 54(6):1345–1358

    Article  Google Scholar 

  36. 36.

    Lin Y-S, Jiang J-Y, Lee S-J (2014) A similarity measure for text classification and clustering. IEEE Trans Knowl Data Eng 26(7):1575–1590

    Article  Google Scholar 

  37. 37.

    Blei DM, Ng AY, Jordan MI (2003) Latent dirichlet allocation. J Mach Learn Res 3:993–1022

    MATH  Google Scholar 

  38. 38.

    Sanborn A, Skryzalin J (2015) Deep learning for semantic similarity, CS224d: deep learning for natural language processing. Stanford University, Stanford

    Google Scholar 

  39. 39.

    Zola P, Cortez P, Ragno C, Brentari E (2019) Social media cross-source and cross-domain sentiment classification. Int J Inf Technol Decis Mak 18(15):1469–1499

    Article  Google Scholar 

  40. 40.

    Tashman L (2000) Out-of-sample tests of forecasting accuracy: an analysis and review. Int Forecast J 16(4):437–450

    Article  Google Scholar 

  41. 41.

    Yamaguchi Y, Takahashi T, Amagasa T, Kitagawa H (2010) Turank: Twitter user ranking based on user-tweet graph analysis. In: International conference on web information systems engineering. Springer, pp 240–253

  42. 42.

    Castillo C, Mendoza M, Poblete B (2011) Information credibility on twitter. In: Proceedings of the 20th international conference on World wide web. ACM, pp 675–684

  43. 43.

    Pal A, Counts S (2011) Identifying topical authorities in microblogs. In: Proceedings of the 4th ACM international conference on Web search and data mining. ACM, pp 45–54

  44. 44.

    Gayo-Avello D (2013) Nepotistic relationships in twitter and their impact on rank prestige algorithms. Inf Process Manag 49(6):1250–1280

    Article  Google Scholar 

  45. 45.

    Ito J, Song J, Toda H, Koike H, Oyama S (2015) Assessment of tweet credibility with LDA features. In: Proceedings of the 24th international conference on world wide web. ACM, pp 953–958

  46. 46.

    Cortez P, Oliveira N, Ferreira JP (2016) Measuring user influence in financial microblogs: experiments using stocktwits data. In: Proceedings of the 6th international conference on web intelligence, mining and semantics. ACM, p 23

  47. 47.

    Eliacik AB, Erdogan N (2018) Influential user weighted sentiment analysis on topic based microblogging community. Expert Syst Appl 92:403–418

    Article  Google Scholar 

  48. 48.

    Alsmadi I, Hoon GK (2019) Term weighting scheme for short-text classification: Twitter corpuses. Neural Comput Appl 31(8):3819–3831

    Article  Google Scholar 

  49. 49.

    Mikolov T, Sutskever I, Chen K, Corrado GS, Dean J (2013) Distributed representations of words and phrases and their compositionality. In: Advances in neural information processing systems, pp 3111–3119

  50. 50.

    Wood-Doughty Z, Andrews N, Dredze M (2018) Convolutions are all you need (for classifying character sequences). In: Proceedings of the 4th workshop on noisy user-generated text, NUT@EMNLP 2018, Brussels, Belgium, November, pp 208–213

  51. 51.

    Manevitz LM, Yousef M (2001) One-class SVMs for document classification. J Mach Learn Res 2:139–154

    MATH  Google Scholar 

  52. 52.

    Senin P (2008) Dynamic time warping algorithm review. Information and Computer Science Department University of Hawaii at Manoa Honolulu, USA 855:1–23

  53. 53.

    Utkin LV, Zaborovsky VS, Lukashin AA, Popov SG, Podolskaja AV (2017) A siamese autoencoder preserving distances for anomaly detection in multi-robot systems. In: International conference on control, artificial intelligence, robotics & optimization (ICCAIRO). IEEE, pp 39–44

  54. 54.

    Xu Y, Jones GJ, Li J, Wang B, Sun C (2007) A study on mutual information-based feature selection for text categorization. J Comput Inf Syst 3(3):1007–1012

    Google Scholar 

  55. 55.

    Oliveira N, Cortez P, Areal N (2016) Stock market sentiment lexicon acquisition using microblogging data and statistical measures. Decis Support Syst 85:62–73

    Article  Google Scholar 

  56. 56.

    Hastie T, Tibshirani R, Friedman J (2008) The elements of statistical learning: data mining, inference, and prediction, 2nd edn. Springer, New York

    MATH  Google Scholar 

  57. 57.

    Goodfellow I, Bengio Y, Courville A, Bengio Y (2016) Deep learning, vol 1. MIT Press, Cambridge

    MATH  Google Scholar 

  58. 58.

    Costa J, Silva C, Antunes M, Ribeiro B (2019) Boosting dynamic ensemble’s performance in twitter. Neural Comput Appl 1–13

  59. 59.

    Batista GE, Prati RC, Monard MC (2004) A study of the behavior of several methods for balancing machine learning training data. ACM SIGKDD Explor Newsl 6(1):20–29

    Article  Google Scholar 

  60. 60.

    Cai J, Lee WS, Teh YW (2007) Improving word sense disambiguation using topic features. In: EMNLP-CoNLL 2007, proceedings of the joint conference on empirical methods in natural language processing and computational natural language learning, Prague, Czech Republic, pp 1015–1023

  61. 61.

    Griffiths TL, Steyvers M (2004) Finding scientific topics. Proc Nat Acad Sci 101(suppl 1):5228–5235

    Article  Google Scholar 

  62. 62.

    Hollander M, Wolfe DA (1999) Nonparametric statistical methods. Wiley, Hoboken

    MATH  Google Scholar 

  63. 63.

    Fawcett T (2006) An introduction to ROC analysis. Pattern Recogn Lett 27(8):861–874

    MathSciNet  Article  Google Scholar 

  64. 64.

    Gonçalves S, Cortez P, Moro S (2019) A deep learning classifier for sentence classification in biomedical and computer science abstracts, Neural Computing and Applications. https://doi.org/10.1007/s00521-019-04334-2

  65. 65.

    Kulkarni R (2018) A million news headlines, Tech. rep., Harvard Dataverse, V2. https://doi.org/10.7910/DVN/SYBGZL

  66. 66.

    Wei Wei, Fan X, Song H, Fan X, Yang J (2018) Imperfect information dynamic stackelberg game based resource allocation using hidden markov for cloud computing. IEEE Trans Serv Comput 11(1):78–89. https://doi.org/10.1109/TSC.2016.2528246

    Article  Google Scholar 

Download references

Acknowledgements

Research carried out with the support of resources of Big and Open Data Innovation Laboratory (BODaI-Lab), University of Brescia, granted by Fondazione Cariplo and Regione Lombardia. We would also like to thank the anonymous reviewers for their helpful suggestions.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Paola Zola.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zola, P., Cortez, P. & Brentari, E. Twitter alloy steel disambiguation and user relevance via one-class and two-class news titles classifiers. Neural Comput & Applic 33, 1245–1260 (2021). https://doi.org/10.1007/s00521-020-04991-8

Download citation

Keywords

  • Text classification
  • User relevance
  • Machine learning
  • Social media analytics