{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,17]],"date-time":"2025-12-17T08:38:31Z","timestamp":1765960711012,"version":"3.48.0"},"reference-count":48,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,5,11]],"date-time":"2025-05-11T00:00:00Z","timestamp":1746921600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,5,11]],"date-time":"2025-05-11T00:00:00Z","timestamp":1746921600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Soc. Netw. Anal. Min."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>In elections worldwide, candidates often resort to negative campaigning due to pressure and the fear of failure. With the rise of social media platforms like Twitter, political discussions are now more accessible than ever. Given the vast amount of data generated, automated systems for detecting negativity in campaigns are crucial to understanding candidate strategies. In this paper, we propose a hybrid model for detecting negativity in campaigns using a two-stage classifier that leverages the strengths of two machine learning models. We collected Persian tweets from 50 political users, including candidates and government officials, and annotated 5,100 tweets published in the year leading up to Iran\u2019s 2021 presidential election. Our model first creates two datasets from the training set for two classifiers by calculating the cosine similarity between tweet embeddings and axis embeddings (the average of positive and negative embeddings). These datasets are then used to train the hybrid model. Our best-performing model (RF-RF) achieved a 79% macro F1 score and 82% weighted F1 score. Applying this model to additional tweets with the help of statistical models, revealed that a candidate\u2019s tweet publication timing does not affect its negativity. Still, the presence of political person names and organization names in tweets is closely linked to negativity.<\/jats:p>","DOI":"10.1007\/s13278-025-01464-5","type":"journal-article","created":{"date-parts":[[2025,5,11]],"date-time":"2025-05-11T07:59:56Z","timestamp":1746950396000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Two-stage classifier for detecting campaign negativity with axis embeddings in persian tweets"],"prefix":"10.1007","volume":"15","author":[{"given":"Fatemeh","family":"Rajabi","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ali","family":"Mohades","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2025,5,11]]},"reference":[{"key":"1464_CR1","doi-asserted-by":"publisher","first-page":"1739","DOI":"10.1007\/s10462-017-9581-3","volume":"52","author":"S Adams","year":"2019","unstructured":"Adams S, Beling PA (2019) A survey of feature selection methods for Gaussian mixture models and hidden Markov models. Artif Intell Rev 52:1739\u20131779","journal-title":"Artif Intell Rev"},{"issue":"1","key":"1464_CR2","first-page":"35","volume":"14","author":"M Babaei","year":"2018","unstructured":"Babaei M, Moradi S, Ghasemi AA (2018) Destruction and negative campaigns in Iran\u2019s presidential elections; causes and contexts. Am Polit Res 14(1):35\u201362","journal-title":"Am Polit Res"},{"key":"1464_CR3","first-page":"1","volume":"15","author":"P Bhattacharjee","year":"2021","unstructured":"Bhattacharjee P, Mitra P (2021) A survey of density based clustering algorithms. Front Comp Sci 15:1\u201327","journal-title":"Front Comp Sci"},{"issue":"4","key":"1464_CR4","first-page":"190","volume":"7","author":"EY Boateng","year":"2019","unstructured":"Boateng EY, Abaye DA (2019) A review of the logistic regression model with emphasis on medical research. J Data Anal Inf Process 7(4):190\u2013207","journal-title":"J Data Anal Inf Process"},{"key":"1464_CR5","first-page":"1","volume":"13","author":"MA Chandra","year":"2021","unstructured":"Chandra MA, Bedi SS (2021) Survey on SVM and their application in image classification. Int J Inf Technol 13:1\u201311","journal-title":"Int J Inf Technol"},{"key":"1464_CR6","unstructured":"Dehghani M, Yazdanparast Z (2023) Political Sentiment Analysis of Persian Tweets Using CNN-LSTM Model. arXiv preprint arXiv:2307.07740"},{"issue":"2","key":"1464_CR7","first-page":"31","volume":"100","author":"EM Dharma","year":"2022","unstructured":"Dharma EM, Lumban Gaol F, Warnars HLHS, Soewito B (2022) The accuracy comparison among Word2Vec, GloVe, and FastText towards convolutional neural network (CNN) text classification. J Theor Appl Inf Technol 100(2):31","journal-title":"J Theor Appl Inf Technol"},{"issue":"1","key":"1464_CR8","doi-asserted-by":"publisher","first-page":"247","DOI":"10.1214\/17-AOS1549","volume":"46","author":"E Dobriban","year":"2018","unstructured":"Dobriban E, Wager S (2018) High-dimensional asymptotics of prediction: ridge regression and classification. Ann Stat 46(1):247\u2013279","journal-title":"Ann Stat"},{"key":"1464_CR9","doi-asserted-by":"publisher","first-page":"3831","DOI":"10.1007\/s11063-021-10528-4","volume":"53","author":"M Farahani","year":"2021","unstructured":"Farahani M, Gharachorloo M, Farahani M, Manthouri M (2021) ParsBERT: transformer-based model for Persian language understanding. Neural Process Lett 53:3831\u20133847","journal-title":"Neural Process Lett"},{"key":"1464_CR10","first-page":"25","volume":"8","author":"H Gohiya","year":"2018","unstructured":"Gohiya H, Lohiya H, Patidar K (2018) A survey of XGBoost system. Int J Adv Technol Eng Res 8:25\u201330","journal-title":"Int J Adv Technol Eng Res"},{"key":"1464_CR11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.15763\/issn.2374-7781.2012.33.0.1-22","volume":"33","author":"M Grossmann","year":"2012","unstructured":"Grossmann M (2012) What (or who) makes campaigns negative? Am Rev Polit 33:1\u201322","journal-title":"Am Rev Polit"},{"issue":"2","key":"1464_CR12","doi-asserted-by":"publisher","first-page":"222","DOI":"10.1177\/1532673X15589613","volume":"44","author":"HJG Hassell","year":"2016","unstructured":"Hassell HJG, Oeltjenbruns KR (2016) When to attack: the trajectory of congressional campaign negativity. Am Polit Res 44(2):222\u2013246","journal-title":"Am Polit Res"},{"key":"1464_CR13","doi-asserted-by":"crossref","unstructured":"Hosseini P, Hosseini P, D A (2020) Broniatowski Content analysis of Persian\/Farsi Tweets during COVID-19 pandemic in Iran using NLP. arXiv preprint arXiv:2005.08400","DOI":"10.18653\/v1\/2020.nlpcovid19-2.26"},{"key":"1464_CR14","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1016\/j.ins.2022.11.139","volume":"622","author":"AM Ikotun","year":"2022","unstructured":"Ikotun AM, Ezugwu AE, Abualigah L, Abuhaija B, Heming J (2022) K-means clustering algorithms: a comprehensive review, variants analysis, and advances in the era of big data. Inf Sci 622:178\u2013210","journal-title":"Inf Sci"},{"issue":"7","key":"1464_CR15","doi-asserted-by":"publisher","first-page":"4359","DOI":"10.1007\/s10994-023-06370-5","volume":"113","author":"E Kebriaei","year":"2024","unstructured":"Kebriaei E, Homayouni A, Faraji R, Razavi A, Shakery A, Faili H, Yaghoobzadeh Y (2024) Persian offensive language detection. Mach Learn 113(7):4359\u20134379","journal-title":"Mach Learn"},{"key":"1464_CR16","doi-asserted-by":"crossref","unstructured":"Khodak M, Saunshi N, Liang Y, Ma T, Stewart B, Arora S (2018) A la carte embedding: Cheap but effective induction of semantic feature vectors. arXiv preprint arXiv:1805.05388","DOI":"10.18653\/v1\/P18-1002"},{"issue":"4","key":"1464_CR17","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1017\/psrm.2022.12","volume":"11","author":"T Kim","year":"2023","unstructured":"Kim T (2023) Violent political rhetoric on Twitter. Polit Sci Res Methods 11(4):673\u2013695","journal-title":"Polit Sci Res Methods"},{"issue":"4","key":"1464_CR18","doi-asserted-by":"publisher","first-page":"150","DOI":"10.3390\/info10040150","volume":"10","author":"K Kowsari","year":"2019","unstructured":"Kowsari K, Jafari Meimandi K, Heidarysafa M, Mendu S, Barnes L, Brown D (2019) Text classification algorithms: a survey. Information 10(4):150","journal-title":"Information"},{"issue":"1","key":"1464_CR19","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1177\/1532673X06292320","volume":"35","author":"TB Krebs","year":"2007","unstructured":"Krebs TB, Holian DB (2007) Competitive positioning, deracialization, and attack speech: a study of negative campaigning in the 2001 Los Angeles mayoral election. Am Polit Res 35(1):123\u2013149","journal-title":"Am Polit Res"},{"issue":"2","key":"1464_CR20","first-page":"1","volume":"13","author":"Q Li","year":"2022","unstructured":"Li Q, Peng H, Li J, Xia C, Yang R, Sun L, Yu PS, He L (2022) A survey on text classification: from traditional to deep learning. ACM Trans Intell Syst Technol (TIST) 13(2):1\u201341","journal-title":"ACM Trans Intell Syst Technol (TIST)"},{"issue":"7","key":"1464_CR21","doi-asserted-by":"publisher","first-page":"100520","DOI":"10.1016\/j.patter.2022.100520","volume":"3","author":"R Liu","year":"2022","unstructured":"Liu R, Li Y, Tao L, Liang D, Zheng H-T (2022) Are we ready for a new paradigm shift? A survey on visual deep MLP. Patterns 3(7):100520","journal-title":"Patterns"},{"issue":"4","key":"1464_CR22","doi-asserted-by":"publisher","first-page":"576","DOI":"10.1177\/1940161220919093","volume":"25","author":"J Maier","year":"2020","unstructured":"Maier J, Nai A (2020) Roaring candidates in the spotlight: campaign negativity, emotions, and media coverage in 107 national elections. Int J Press\/Polit 25(4):576\u2013606","journal-title":"Int J Press\/Polit"},{"issue":"2","key":"1464_CR23","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1177\/0192512121994512","volume":"44","author":"J Maier","year":"2023","unstructured":"Maier J, Nai A (2023) Mapping the drivers of negative campaigning: insights from a candidate survey. Int Polit Sci Rev 44(2):195\u2013211","journal-title":"Int Polit Sci Rev"},{"key":"1464_CR24","volume-title":"The positive case for negative campaigning","author":"K Mattes","year":"2020","unstructured":"Mattes K, Redlawsk DP (2020) The positive case for negative campaigning. University of Chicago Press, Chicago"},{"key":"1464_CR25","first-page":"9540","volume":"33","author":"F Mignacco","year":"2020","unstructured":"Mignacco F, Krzakala F, Urbani P, Zdeborov\u00e1 L (2020) Dynamical mean-field theory for stochastic gradient descent in Gaussian mixture classification. Adv Neural Inf Process Syst 33:9540\u20139550","journal-title":"Adv Neural Inf Process Syst"},{"issue":"3","key":"1464_CR26","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3439726","volume":"54","author":"S Minaee","year":"2021","unstructured":"Minaee S, Kalchbrenner N, Cambria E, Nikzad N, Chenaghlu M, Gao J (2021) Deep learning-based text classification: a comprehensive review. ACM Comput Surv (CSUR) 54(3):1\u201340","journal-title":"ACM Comput Surv (CSUR)"},{"issue":"3","key":"1464_CR27","doi-asserted-by":"publisher","first-page":"430","DOI":"10.1017\/gov.2018.32","volume":"55","author":"A Nai","year":"2020","unstructured":"Nai A (2020) Going negative, worldwide: towards a general understanding of determinants and targets of negative campaigning. Gov Oppos 55(3):430\u2013455","journal-title":"Gov Oppos"},{"issue":"2","key":"1464_CR28","first-page":"219","volume":"20","author":"A Nai","year":"2021","unstructured":"Nai A (2021) Fear and loathing in populist campaigns? Comparing the communication style of populists and non-populists in elections worldwide. J Polit Mark 20(2):219\u2013250","journal-title":"J Polit Mark"},{"key":"1464_CR29","doi-asserted-by":"publisher","first-page":"688","DOI":"10.1016\/j.ins.2022.06.027","volume":"607","author":"T Ouyang","year":"2022","unstructured":"Ouyang T, Shen X (2022) Online structural clustering based on DBSCAN extension with granular descriptors. Inf Sci 607:688\u2013704","journal-title":"Inf Sci"},{"key":"1464_CR30","doi-asserted-by":"crossref","unstructured":"Pareek J, Jacob J (2021) Data compression and visualization using PCA and T-SNE. In Advances in Information Communication Technology and Computing: Proceedings of AICTC 2019, 327\u2013337. Springer","DOI":"10.1007\/978-981-15-5421-6_34"},{"issue":"1","key":"1464_CR31","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1177\/106591290505800104","volume":"58","author":"DAM Peterson","year":"2005","unstructured":"Peterson DAM, Djupe PA (2005) When primary campaigns go negative: the determinants of campaign negativity. Polit Res Q 58(1):45\u201354","journal-title":"Polit Res Q"},{"issue":"3","key":"1464_CR32","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1177\/1532673X211055676","volume":"50","author":"V Petkevic","year":"2022","unstructured":"Petkevic V, Nai A (2022) Political attacks in 280 characters or less: a new tool for the automated classification of campaign negativity on social media. Am Polit Res 50(3):279\u2013302","journal-title":"Am Polit Res"},{"key":"1464_CR33","doi-asserted-by":"crossref","unstructured":"Pires T, Schlinger E, Garrette D (2019) How multilingual is multilingual BERT? arXiv preprint arXiv:1906.01502","DOI":"10.18653\/v1\/P19-1493"},{"issue":"4","key":"1464_CR34","doi-asserted-by":"publisher","first-page":"882","DOI":"10.1007\/s12559-021-09886-x","volume":"13","author":"Z Rajabi","year":"2021","unstructured":"Rajabi Z, Valavi M (2021) A survey on sentiment analysis in Persian: a comprehensive system perspective covering challenges and advances in resources and methods. Cogn Comput 13(4):882\u2013902","journal-title":"Cogn Comput"},{"issue":"1","key":"1464_CR35","doi-asserted-by":"publisher","first-page":"012090","DOI":"10.1088\/1757-899X\/725\/1\/012090","volume":"725","author":"F Ramadhani","year":"2020","unstructured":"Ramadhani F, Zarlis M, Suwilo S (2020) Improve BIRCH algorithm for big data clustering. IOP Conf Ser: Mater Sci Eng 725(1):012090","journal-title":"IOP Conf Ser: Mater Sci Eng"},{"issue":"2","key":"1464_CR36","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1007\/s00357-020-09377-y","volume":"38","author":"N Randriamihamison","year":"2021","unstructured":"Randriamihamison N, Vialaneix N, Neuvial P (2021) Applicability and interpretability of Ward\u2019s hierarchical agglomerative clustering with or without contiguity constraints. J Classif 38(2):363\u2013389","journal-title":"J Classif"},{"issue":"3","key":"1464_CR37","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3178582","volume":"51","author":"PAA Resende","year":"2018","unstructured":"Resende PAA, Drummond AC (2018) A survey of random forest based methods for intrusion detection systems. ACM Comput Surv (CSUR) 51(3):1\u201336","journal-title":"ACM Comput Surv (CSUR)"},{"key":"1464_CR38","unstructured":"Sanh V, Debut L, Chaumond J, Wolf T (2019) DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108"},{"key":"1464_CR39","unstructured":"Schweitzer E J (2009) Attack politics on the internet: Comparing German and American E-campaigns. APSA 2009 Toronto Meeting Paper"},{"key":"1464_CR40","doi-asserted-by":"crossref","unstructured":"Sha\u2019Abani M N A H, Fuad N, Jamal N, Ismail M F (2020) kNN and SVM classification for EEG: a review. In InECCE2019: Proceedings of the 5th International Conference on Electrical, Control & Computer Engineering, Kuantan, Pahang, Malaysia, 29th July 2019, 555\u2013565. Springer","DOI":"10.1007\/978-981-15-2317-5_47"},{"key":"1464_CR41","first-page":"65","volume":"11","author":"M Shamsfard","year":"2011","unstructured":"Shamsfard M (2011) Challenges and open problems in Persian text processing. Proc LTC 11:65\u201369","journal-title":"Proc LTC"},{"key":"1464_CR42","doi-asserted-by":"publisher","first-page":"105942","DOI":"10.1016\/j.asoc.2019.105942","volume":"86","author":"R Sun","year":"2020","unstructured":"Sun R, Wang G, Zhang W, Hsu L-T, Ochieng WY (2020) A gradient boosting decision tree based GPS signal reception classification algorithm. Appl Soft Comput 86:105942","journal-title":"Appl Soft Comput"},{"issue":"1","key":"1464_CR43","first-page":"168","volume":"17","author":"A Tharwat","year":"2020","unstructured":"Tharwat A (2020) Classification assessment methods. Appl Comput Inf 17(1):168\u2013192","journal-title":"Appl Comput Inf"},{"key":"1464_CR44","doi-asserted-by":"crossref","unstructured":"Thavareesan S, Mahesan S (2020) Sentiment lexicon expansion using Word2vec and fastText for sentiment prediction in Tamil texts. In 2020 Moratuwa Engineering Research Conference (MERCon), 272\u2013276. IEEE","DOI":"10.1109\/MERCon50084.2020.9185369"},{"issue":"3","key":"1464_CR45","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1177\/1940161219843404","volume":"24","author":"AS Walter","year":"2019","unstructured":"Walter AS, Van der Eijk C (2019) Measures of campaign negativity: comparing approaches and eliminating partisan bias. Int J Press\/Polit 24(3):363\u2013382","journal-title":"Int J Press\/Polit"},{"issue":"3","key":"1464_CR46","doi-asserted-by":"publisher","first-page":"2277","DOI":"10.1007\/s00500-020-05297-6","volume":"25","author":"I Wickramasinghe","year":"2021","unstructured":"Wickramasinghe I, Kalutarage H (2021) Naive Bayes: applications, variations, and vulnerabilities: a review of the literature with code snippets for implementation. Soft Comput 25(3):2277\u20132293","journal-title":"Soft Comput"},{"key":"1464_CR47","doi-asserted-by":"crossref","unstructured":"Wijayanto U W, Sarno R (2018) An experimental study of supervised sentiment analysis using Gaussian Na\u00efve Bayes. In 2018 International Seminar on Application for Technology of Information and Communication, 476\u2013481. IEEE","DOI":"10.1109\/ISEMANTIC.2018.8549788"},{"key":"1464_CR48","doi-asserted-by":"crossref","unstructured":"Zhou K, Ethayarajh K, Card D, Jurafsky D (2022) Problems with cosine as a measure of embedding similarity for high frequency words. arXiv preprint arXiv:2205.05092","DOI":"10.18653\/v1\/2022.acl-short.45"}],"container-title":["Social Network Analysis and Mining"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13278-025-01464-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13278-025-01464-5\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13278-025-01464-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,17]],"date-time":"2025-12-17T08:28:44Z","timestamp":1765960124000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13278-025-01464-5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,11]]},"references-count":48,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["1464"],"URL":"https:\/\/doi.org\/10.1007\/s13278-025-01464-5","relation":{},"ISSN":["1869-5469"],"issn-type":[{"type":"electronic","value":"1869-5469"}],"subject":[],"published":{"date-parts":[[2025,5,11]]},"assertion":[{"value":"11 February 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 February 2025","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 April 2025","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 May 2025","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"51"}}