Human being is a “Homo Sapiens” which means a “wise-man” having some specific capabilities such as thinking, reasoning, acting, and learning and to behave (Lee and Cho, 2020). Besides, these human capabilities make them able to understand their environment and act accordingly. Reciprocally, this ability of humans is explained by a word “intelligence” which is borrowed from Latin word which means “To Understand” (Kumar and Sha, 2017). It was 1956, when the word “Artificial Intelligence” (AI) was first time utter in a conference (Russell and Norvig, 1995; Martínez-López and Casillas, 2013). At the initial stage the question was to bring human brain ability into a machine system which could behave like human mind. Although some of the technologies like home appliances and motors have claimed to be intelligent but these devices do not have capability of reasoning, perception, action and behavior (Freier, 2016). Additionally, the concept of AI was the beginning of robotics (Perez et al., 2018). Later on, several fields, such as business, medical science, agriculture, and construction, have rendered the services of AI to serve their target audience (Ribeiro et al., 2021).

It is not possible for business world to ignore the decisiveness of AI to target their customers. There are multiple tasks in business, which need valuable and automated solution for their success (Negnevitsky and Intelligence, 2005). To forecast future, business executives need ad-hoc intelligence system (Martínez-López and Casillas, 2009; Casillas et al., 2012; Chen et al., 2021). In addition to the potentialities, AI assist companies in strategic intelligence such as managing customer relationship, positioning products, segmenting the market, designing personalized advertisements, and product-harm crisis (Martínez-López and Casillas, 2013; Dubey et al., 2021). To understand consumer is very complex because there are numerous channels and forms through which they express their needs and attitudes (Kietzmann et al., 2018). Companies demand technological innovations to conduct advertising tasks (Qin and Jiang, 2019). The transformation of business model from physical to online has also changed the advertising method. Therefore, for effective communication and marketing, companies are paying much attention on online brand communality engagement OCBE (de Almeida et al., 2018). Consumer's unstructured and freely written online comments at various blogs like Twitters, Facebook and other social media channels are very valuable for companies in improving their products. The application of AI in advertising has emerged the concept of “smart advertising” (Vakratsas and Wang, 2020). AI technology has restructured the advertising process (Qin and Jiang, 2019). It is not confined to design advertisement but to understand the consumers' emotions and cognitions toward advertisements contents (Abbas et al., 2018).

The advertising cost is increasing worldwide. It has been observed that on global digital advertisement, the estimated cost has risen to 17.6% (Lee and Cho, 2020). Therefore, to cut short the ineffective advertisement there is a need to understand consumer attention, cognition, and emotion toward advertisements. Companies are trying to figure out solutions for effective advertisements by using big data like social media comments. Aforementioned research has suggested investigating the consumer insight for designing consumer oriented advertisement (Qin and Jiang, 2019). Therefore, the consumer life-style and behavior is imperatively essential to be understood (Li et al., 2021). Whereas, gaining the understanding about the consumer life-style and behavior is almost impossible without the use of innovative technological tool like AI (Di Vaio et al., 2020). Owing to the increasing cost of advertisement and rapid failure of advertising contents have increased companies' unnecessary financial budget. Moreover, to get competitive edge the consumer insights is inevitable. As Shrivastava (2014) stated that 1 of the extensively accepted source of information for the consumers is advertisement, that's why choice of media is pivotal for the marketers.

Additionally, consumers' online feedback exhibits their true representation as it has based on their experiences. Eventually, the objective of this study is to understand consumer attention, cognition and emotion toward advertisements by using big data sources. Extracting the online consumer content helps the decision makers to analyze the customers' sentiments and to haul out the required information. AI can help in data gathering and affective computing for sentimental analysis, which ultimately helps marketers to design the effective customer satisfaction strategies but also to manage their employee engagement (Khrais, 2020).

A profound revolution has been observed in advertising. Marketers must face and embrace the new changes while conducting advertisement campaign to reach and communicate product features to customers. The composition of advertisement in digital era has transformed and advertisers are shifted from electronic and print media to social media. Advertising researchers and practitioners have raised a question regarding changing nature of advertisement “will we continue to follow the same pathways and trace the same footprints of those who have gone before us, or will we set off on a new path, embracing change and acknowledging that new approaches are needed?” (Schultz and Schultz, 2016). Advertiser generates various appeals while targeting customers such as cognitive, emotion, and attention. The nature of product defines which appeal is suitable for delivering product information. The extraction of consumers' emotion, cognition, and attention from their online comments is not an easy task. These cognitions and emotions are drivers to information and message processing that leads to formation of attitude (Zaki et al., 2021). The availability of big data in the form of consumer feedback pertaining to the companies' advertisement must utilize for consumer insight. To manage unstructured online data, there is need of systematically programed software which can read human language and transformed it into usable form.

The current model has underpinned with cognitive theory. According to Jean Piaget's theory of cognitive, the cognitive theory is characterized by “their focus on the idea that how and what people think leads to the arousal of emotions and that certain thoughts and beliefs lead to disturbed emotions and behaviors and others lead to healthy emotions and adaptive behavior”. The individual share their thoughts and feeling with complete freedom of expression at their tweeters accounts. These comments regarding any object facilitate scholars to evaluate individual's positive and negative opinions. Therefore, the objective of the current study is to evaluate individual's emotions, cognition and attention for purchasing a mobile product, which is facing defamation issue. The selection of such products has required a strong emotional, cognitive, and attention.

In consumer behavior the qualitative research method has become essential because of its classical advantages in dealing with big data and data mining (Petrescu and Lauer, 2017). Moreover, the numerous sources of big data have the data collection processes. For consumers' insight, researchers and practitioners are more interested into the techniques, which can handle big data rather using traditional close-ended questionnaires. Social media and companies' official blogs are not only the source of big data but also generate neutral opinions with freedom of expressions.

This research has utilized the Shapiro's (1994) Think-aloud procedure to study the consumer insights. In this procedure, individuals have two ways of expressing their opinions; dialogue and writing. Consumers usually use different mediums to express their sentiments such as Facebook, twitter, etc. For this research, twitter's data set was taken into account for analysis. Aforementioned studies witnessed that twitters and Facebook are convenient, easy, and cheap social media methods for generating massive word of mouth for both positive and negative (Javornik et al., 2020). Therefore, authors of the study involved twits data for analyzing consumer opinions. Besides, authors have evaluated consumer's twits for Samsung Galaxy. The purpose to select Samsung Galaxy was an incident related to Samsung Galaxy Note 7 when in 2016 news spread in the market that Samsung Galaxy Note 7's battery exploded and US flights banned the product (Krisher, 2019). This was the product-harm crisis situation. Later, Samsung has fixed the problem and re-launch the product and also introduced Galaxy Note 8. Nevertheless there was a fear which has prevailed in the market for a time being. Now authors of the study interested to understand consumers' opinions about Samsung Galaxy with respect to product features, brand loyalty, and product image by using individual's sentiment's proxy like emotion, cognition, and attention. There are some official websites which maintained twitters' dataset formally and used for research analysis like GitHub and Data, World, etc. Authors have divided the twits into two sections like international tweets which comprised 30877 and Pakistani consumers' tweets dataset which was 26834. Tweets were analyzed using natural language processing software Nvivo 12 and generated word clouds. In this research authors' focused on the data available from 2016 to 2018. The data was further classified into two categories (i) 2016 to 2017 and (ii) 2017 to 2018 to have a better understanding about the consumer's changing sentiments.

The objective of the study was to analyze consumers' unstructured opinions to get their insights toward an advertisement. Author of the study has employed twitters' dataset regarding Samsung Galaxy. Moreover, for the data analysis, natural language process (NLP) was used named Nvivo 12. The tweets were comprised of consumers' 3 different sentiments like emotion, cognition and attention. Owing to the increasing competition companies have very limited options to get competitive edge even in highly innovative products like cell phones. Furthermore, for value addition companies should understand consumer's preferences. Companies stimulate consumer's need with advertisements and these advertisements are the source to target consumer's emotion, grab consumer's attention and strike consumer's cognitive behavior. To collect consumer's feedback with structured and self-administered questionnaire make consumer confined in limited options. Therefore, the involvement of big data and algorithm to evaluate consumers' extensive opinions make significant difference as compared to traditional method of data collection. The involvement of big data and algorithm are the techniques of artificial intelligence to transform the unstructured and meaningless data into meaningful themes. Artificial intelligence has shifted the method to advertise and guide consumers. Moreover, for consumer-generated data mining various techniques of artificial intelligence are play vital role. Researchers and practitioners are suggesting using big data for designing consumer oriented advertisements. There is a need of ad- hock intelligence system, which can assist individual to evaluate and interpret consumer's post-purchase unstructured expression.

The interpretation of consumers' responses on Facebook and twitters can cause the success of advertisement. These social media platforms are the best source for consumer to express their post purchase opinions. The results of current study have unfolded the fact that consumer's share their emotions, attention and cognitive behavior in their tweets which are prominently appear in word cloud. These online opinions are also an electronic word of mouth (e-WOM) and play as an advertisement for potential customers (Araujo et al., 2017; Miremadi et al., 2021). Aforementioned studies have suggested that to succeed in market and to design effective advertisement the advertiser should frequently monitor consumers' electronic unstructured opinions (Colliander et al., 2015; Craig et al., 2015). The unstructured consumer's data is the window of opportunity for advertisers, researchers, practitioners, and policy makers to target consumers with customized advertisement (Chu and Kim, 2018). In addition to machine leaning, assisting advertisers to collect consumer data from multiple sources and mine them to deliver on spot consumer insights.

The current study has some limitations and author has suggestions for future research. The tweets dataset of the current study was taken from GitHub.com and dataset. In future it is better to conduct an in depth interviews by involving consumers who are using various cell phones. The data collected from in depth interviews should be compare with online posted tweets. Tweets also have some limitations like word count and most often consumer are unable to express their sentiments in detail. Correlation can also be used to measure and describe the relationship between opinions and factors affecting these opinions. Because, it can also be used for prediction as when 2 variables are correlated, then 1 variable can be used to predict the other variable through artificial intelligence and machine learning. This can give us a more accurate analysis of the twitter data and its values. Moreover, in tweets dataset it is not possible to differentiate that which users are sharing their own experience and which are just endorsing others opinions. Furthermore, although NLP Nvivo 12 generates themes from unstructured and meaningless sentences nevertheless in future researchers should involve more sophisticated AI software for data analysis.

The concept of artificial intelligence has inculcated in computer science and medical science. Additionally, the artificial intelligence is also serving to business while taking various recruitment decisions, marketing decisions and for reviewing advertisement feedbacks. Owing to the presence of digital environment customers are very much accustomed with and experiencing digital advertisement. Moreover, the current study is a small contribution in existing literature which is opening new academic venue for AI significance in consumer' insights.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

HS provided the services for the proofreading as well as analysis of the article. MZ wrote the introduction and literature of the article. NZ wrote the methodology and discussion part of the article. All authors contributed to the article and approved the submitted version.

This work was funded by Business School, Shandong Jianzhu University, China to HS.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.