Improving sentiment analysis for twitter data by handling negation rules in the Serbian language

Ljajić, Adela; Marovac, Ulfeta

doi:10.2298/csis180122013l

Cited by 16 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They used datasets in 13 European languages, one of which was Serbian, and three sentiment classes—positive, negative, and neutral—but they found the Serbian data to suffer from quite low inter-annotator agreements. More recently, Ljajić and Marovac [ 39 ] evaluated different ways of handling negation in Serbian, using a corpus of tweets divided into the positive, negative, and neutral class, but they did not discuss what—if any—annotation guidelines they followed and the resulting dataset was not made public.…”

Section: Related Workmentioning

confidence: 99%

“…Although rather basic, this method can be useful for languages and domains in which an adequate syntactic parser cannot be found. Since this is the case for the informal register of Serbian, which is predominant in our dataset, and since previous experiments on review-length documents [ 36 ] and tweets in Serbian [ 39 ] showed this approach to be beneficial, we explore its usefulness here, as well. In particular, we experiment with different scopes of negation marking, ranging from a single word after a negation to all the words between a negation and a punctuation symbol.…”

Section: Sentiment Analysismentioning

confidence: 99%

“…In particular, we experiment with different scopes of negation marking, ranging from a single word after a negation to all the words between a negation and a punctuation symbol. In doing so, we follow the negation scope rules and patterns proposed by Ljajić and Marovac [ 39 ].…”

Section: Sentiment Analysismentioning

confidence: 99%

See 2 more Smart Citations

A versatile framework for resource-limited sentiment articulation, annotation, and analysis of short texts

2020

View full text Add to dashboard Cite

Choosing a comprehensive and cost-effective way of articulating and annotating the sentiment of a text is not a trivial task, particularly when dealing with short texts, in which sentiment can be expressed through a wide variety of linguistic and rhetorical phenomena. This problem is especially conspicuous in resource-limited settings and languages, where design options are restricted either in terms of manpower and financial means required to produce appropriate sentiment analysis resources, or in terms of available language tools, or both. In this paper, we present a versatile approach to addressing this issue, based on multiple interpretations of sentiment labels that encode information regarding the polarity, subjectivity, and ambiguity of a text, as well as the presence of sarcasm or a mixture of sentiments. We demonstrate its use on Serbian, a resource-limited language, via the creation of a main sentiment analysis dataset focused on movie comments, and two smaller datasets belonging to the movie and book domains. In addition to measuring the quality of the annotation process, we propose a novel metric to validate its cost-effectiveness. Finally, the practicality of our approach is further validated by training, evaluating, and determining the optimal configurations of several different kinds of machine-learning models on a range of sentiment classification tasks using the produced dataset.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Sentiment Analysismentioning

confidence: 99%

See 1 more Smart Citation

A versatile framework for resource-limited sentiment articulation, annotation, and analysis of short texts

2020

View full text Add to dashboard Cite

show abstract

“…Labeling the range of negation is of importance in medical records. For the purpose of denoting the negation symbol, a set of negation symbols together with a set of rules proposed in a given paper was used [31].…”

Section: Non-medical Resourcesmentioning

confidence: 99%

Automated labeling of terms in medical reports in Serbian

2020

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Nowadays, many electronic health reports (EHRs) are stored daily. They consist of the structured part and of an unstructured section written in natural language. Due to the limited time for medical examination, EHRs are short reports which often contain errors and abbreviations. Therefore it is a challenge to process an EHR and extract knowledge from this part of the text for different purposes. This paper compares the results of three proposed methods for automatic labeling of medical terms in unstructured parts of EHRs. All words are categorized as words within the medical domain (symptoms, diagnoses, therapies, anatomy, specialties etc.) and those beyond the medical domain (numbers, places, stop words etc.). The first method is based on dictionaries of medical terms, the second on the training set, and the third on the training set and rules. The results of application of different methodologies to reduce a word to its basic form (pure, prefix, stem) are given for each of the methods. The paper shows that in labeling medical terms, the methods based on medical dictionaries (diagnosis, symptoms, medications etc.) do not produce best results, therefore it is better to use manually annotated part of the data set as a model. A significant number of words (17.36%) in medical reports are abbreviations and errors, so for better results, we should focus on creating rules to solve this problem. Better results are obtained for supervised methods compared to the dictionary-based method (with relative improvement of 42.82%). The inclusion of the algorithm for processing errors and abbreviations increased the results (with a relative improvement of 4.21%) and gave the largest F1-measure (0.9082). The advantage of the proposed method is that the use of rules for processing errors and abbreviations provides good results regardless of how the word is reduced to its basic form.

show abstract

“…An overview of the methods for analyzing sentiment data in Twitter is given in [33]. In the paper [34] the authors examined how the treatment of negation impacts the sentiment of tweets in the Serbian language. A study on social influence produced by the contents published on Facebook was presented in paper [35].…”

Section: Related Workmentioning

confidence: 99%

Symbols: Software for Social Network Analysis

et al. 2019

View full text Add to dashboard Cite

The unique possibilities of the online social networks such as real-time data access, knowledge of users’ changing preferences and access to their statuses provide the possibility for innovation in the analysis of people’s behavior and opinions, when compared to classical offline methods. Literature review shows lack of studies about the use of public Facebook data in Serbia for the improvement of different product sale, political or promotional campaigns, recommender systems, etc. In this paper, we present the way how data from Facebook can be collected in order to gain insight into the individuals’ preferences and statuses, as well as their connection to a company's fan pages. In particular, we present data collection framework – Symbols – used for collecting individual specific data. The framework stores data into local database and involves a module for graph and content-based analysis of these data. The proposed framework for social network analysis can be used as a decision-making system in users’ preferences implementation thus creating a space for business improvements in various areas.

show abstract

Improving sentiment analysis for twitter data by handling negation rules in the Serbian language

Cited by 16 publications

References 20 publications

A versatile framework for resource-limited sentiment articulation, annotation, and analysis of short texts

A versatile framework for resource-limited sentiment articulation, annotation, and analysis of short texts

Automated labeling of terms in medical reports in Serbian

Symbols: Software for Social Network Analysis

Contact Info

Product

Resources

About