Research Article A Generalized Method for Sentiment Analysis across Different Sources Abubakar M. Ashir Department of Computer Engineering, Tishk International University, Erbil, Iraq CorrespondenceshouldbeaddressedtoAbubakarM.Ashir;abubakar.ashir@tiu.edu.iq Received 2 October 2021; Revised 27 November 2021; Accepted 5 December 2021; Published 18 December 2021 AcademicEditor:FrancescoRundo Copyright©2021AbubakarM.Ashir.isisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Sentimentanalysisiswidelyusedinavarietyofapplicationssuchasonlineopiniongatheringforpolicydirectivesingovernment, monitoring of customers, and staff satisfactions in corporate bodies, in politics and security structures for public tension monitoring,andsoon.Inrecenttimes,thefieldmetwithnewsetofchallengeswherenewalgorithmshavetocontendwithhighly unstructuredsourcesforsentimentexpressionsemanatingfromonlinesocialmediafora.Inthisstudy,aruleandlexical-based procedure is proposed together with unsupervised machine learning to implement sentiment analysis with an improved generalizationabilityacrossdifferentsources.Todealwithsourcesdevoidofsyntacticandgrammaticalstructure,theapproach incorporatesaruled-basedtechniqueforemoticondetection,wordcontractionexpansion,noiseremoval,andlexicon-basedtext preprocessingusinglexicalfeaturessuchaspartofspeech(POS),stopwords,andlemmatizationforlocalcontextanalysis.Atext isbrokenintonumberoftokenswitheachrepresentingasentenceandthenlexicon-dependentfeaturesareextractedfromeach token.efeaturesaremergedtogetherusingacombiningfunctionforagiventextbeforebeingusedtotrainamachinelearning classifier. e proposed combining functions leverage on averaging and information gain concepts. Experimental results with different machine leaning classifiers indicate that improved performance with great deal of generalization capacity across both structuredandnonstructuredsourcescanberealized.efindingshowsthatcarefullydesignedlexicalfeaturesreinforcelearning processinunsupervisedlearningmorethanusingwordembeddingsaloneasthefeatures.Obtainedexperimentalresultsfrom movie review dataset (recall � 74.9%, precision � 70.9%, F1-score � 72.9%, and accuracy � 72.0%) and twitter samples’ datasets (recall � 93.4%, precision � 89.5%, F1-score � 91.4%, and accuracy � 91.1%) show the efficacy of the proposed approach in comparison with other state-of-the-art research studies. 1. Introduction Sentiment analysis is a part of natural language processing (NLP) which receives tremendous attention in recent his- tory. is may not be unconnected to the availability of social media platforms, big data storage, increased Internet connectivity, accessibility, and unending desire by big businessandgovernmentstounderstandpeople’sopinions forpolicyconceptualizationsandmonitoring.Atthebackof thisboomistherecentbreakthroughinmachineanddeep learning algorithms leading to an astronomical improve- ment in performance of NLP tasks. Sentiment analysis crisscrosses subfields of computational linguistic and in- formation retrieval. In general context, the major task in sentiment analysis has to do with tagging a given text accordingtoexpressedopinionwhichusuallyinvolvesthree tasks: (i) determine objectivity of a text (i.e., subjective or objective),(ii)determinethepolarityofasubjectivetext(i.e., positiveornegative),and(iii)determinethestrengthofthe subjectivetext[1].erearetwomajorapproachesthatexist in the literature for sentiment analysis: lexicon-based and machinelearning-basedapproach.Eachoftheseapproaches hastheirbenefitsanddrawbacks.Lexicon-basedapproachis arule-basedmethodwhichemployscomputingsentiments by considering the semantic orientation of the words or phrasesinthetext[1].isimpliestheuseofadictionaryof words which are tagged with lexical features such as sen- timent polarity orientation, part of speech (POS), and glosses.Infact,theapproachrepresentsapieceofwordasa tokenorabagofwordswheresemanticorientationofeach Hindawi Applied Computational Intelligence and So Computing Volume 2021, Article ID 2529984, 8 pages https://doi.org/10.1155/2021/2529984