Using big data to understand public opinion: How is artificial intelligence applied in text data research?
Author: uMax Data Team (Angus Cheong, Zao Ying, Cao Wen Yuan)
In an era when everyone is talking about big data and artificial intelligence, can the future of social science research go hand in hand with these new technological developments?How can we combine artificial intelligence and social science research to gain insights into public opinion using big data?
Social science is the science that explores human society and its development laws. This field involves disciplines such as philosophy, economics, law, politics, sociology, history, literature, and art. With the advent of the big data era，Artificial Intelligence、Machine Learning、Deep LearningIt has brought new opportunities and perspectives to social science research, but at the same time, the relatively weak knowledge of technology and algorithms mastered by researchers in the social science field has made them reluctant to apply technologies such as artificial intelligence.
What is Artificial Intelligence？Artificial intelligence mainly refers to machines performing tasks in a way that imitates human intelligence.[1] Specifically, artificial intelligence can be understood from three levels.
The first level is more general and refers to the ability of machines to perform tasks that we usually understand (human-like understanding). [2] The second level integrates multiple human-like capabilities, that is, machines have the ability to sense, understand, act, and learn like humans. [3]
The third level rises to the ability to recognize, judge and solve problems. Artificial intelligence shows cognitive and executive abilities similar to those of humans, emphasizing that artificial intelligence is a complex technological application in which machines can demonstrate human cognitive functions such as learning, analysis and problem solving. [4]
Generally speaking, artificial intelligence mainly focuses on the exploration and practice of human-like perception, cognition and judgment abilities. Depending on whether the machine has autonomous consciousness, it can be divided into1 Strong artificial intelligence with autonomous consciousness 2 Weak artificial intelligence without autonomous consciousness[5] Weak artificial intelligence mainly simulates certain specific skills of humans and intelligently handles problems in certain specific scenarios and applications. Practical application areas include, for example, speech recognition, image and face recognition, natural language processing, data retrieval, autonomous driving, intelligent control robots, etc.
The development of artificial intelligence is currently still at the weak artificial intelligence stage, and is striving to make a breakthrough into strong artificial intelligence with autonomous consciousness.
in social science research area，Artificial intelligence is often used in data analysis and processing, especially in the mining of meaning and value insights of text data.Including news reports, social network information, historical archives, interview texts, literature, policy documents, etc. How to use artificial intelligence to understand people’s behavior and thoughts from text?
The application of artificial intelligence in the mining and analysis of text big data focuses on media monitoring and trend forecasting.Taking research and practice in the field of public opinion as an example, many current applications are in the stage of collecting intelligence, which is the perception level of artificial intelligence. For example, when using machines to obtain data, it is necessary to consider the issue of data coverage, specifically whether the data is complete, representative, and of data qualityPerception is related. Cognition, which is equivalent to the machine's intelligent automatic classification and analysis through understanding of natural language text. In practical terms, it is about how to measure and derive meaning and insights from the text. judgment，It is equivalent to the decisions and actions made after understanding the text. In other words, how to interpret, analyze, and explore research findings to help users make correct judgments and provide guidance and reference for subsequent actions.These three steps are the issues that need to be considered when using artificial intelligence to assist in text data mining and analysis. These are also the three major challenges encountered in the current process of text big data mining and analysis. For detailed instructions, please refer to: https://mp.weixin.qq.com/s/pnd4UzAQmCudVF0tU6C-fw
Currently, the application of artificial intelligence in text big data mining is still relatively preliminary.
Taking the public opinion system as an example, the current focus is on describing the results of KPIs (as shown in Figure 1), such as data sources, content classification, number of likes, number of replies, amount of sharing, popularity, sentiment analysis, and emotional analysis. These are all part of machine perception, but currently this analytical capability is still relatively superficial. If you need to find insights that can help with decision-making, further in-depth exploration and analysis are still needed.
Current technology-driven analysis often produces programmed, homogenized automated charts as its main output, but it limits our imagination, interpretation and judgment, thus limiting our ability to make more insightful discoveries. Therefore, from perception, cognition to judgment, what is important is not the visualization results we see, but the confidence these results can bring us in decision-making and judgment. This requires that everything from establishing the database to setting up the analysis framework, from measurement to analysis, should be controlled by us "humans". It is the analysts who decide what the machine shows us, not the machines that decide what we see.
Information Technology—>Social SciencesThe text big data mining process goes through several different stages. Initially, the information was categorized by searching electronic newsletters. Up to now, it is the practice of public opinion monitoring and brand listening. From symbols/signals to classified data to data visualization, there are already many practices and applications. Currently, most public opinion analysis tools remain at this stage, constantly researching systems or programs that use machines for automated analysis.
However, if we want to conduct analysis and mining after monitoring, we need to have a new understanding of text big data mining, that is, to shift from the perspective of information technology to the perspective of social science. In other words,Information technology should be used to assist social science thinking and analytical methods.
How to realize this human-machine integration mechanism and combine big data technology with social science research methods? The solution provided by Source Big Data is to use artificial intelligence (AI) combined with social science research methods, big data technology and professional natural language processing technology (NLP), including Mandarin, Cantonese and English, to extract insights from the massive amount of chaotic online information collected from social media and the Internet, and provide decision-making references for customers.
Through the source big data solution, analysts can conveniently carry out research design, execution, and result presentation in one stop. Source Big Data's products use a complete set of scientific and systematic big data technologies to assist online content analysis methods, and are designed to be flexible and easy to use, allowing users from different backgrounds to use their industry wisdom and domain expertise to find truly useful insights, including policy research, corporate image, market analysis, public relations crisis, product research, etc., which can all be used appropriately.
In addition, uMax Data analysis platform is flexible and resilient. In addition to intelligent visualization charts that extract accurate insights, users can also create their own charts to meet different research and analysis needs. The platform's analysis results can be verified at any time and traced back to the original text layer by layer, and charts can be directly downloaded and data updated, which is extremely transparent.
References
[1] Marr, B. (2016). What is the difference between artificial intelligence and machine learning? Forbes. Retrieved from: https://www.forbes.com/sites/bernardmarr/2016/12/06/what-is-the-difference-between-artificial-intelligence-and-machine-learning/#66fdb6862742.
[2] Knowledge@Wharton (2018). Vishal Sikka: Why AI needs a broader, more realistic approach. Retrieved fromhttp://knowledge.wharton.upenn.edu/article/ai-needsbroader- realistic-approach/.
[3] Daugherty, P., Carrel-Billiard, M., & Biltz, M. (2018). Accenture technology vision 2018. Retrieved from. Intelligent Enterprise Unleashed. Accenturehttps://www. accenture.com/t00010101T000000Zw/nz-en/_acnmedia/Accenture/next-gen-7/tech-vision-2018/pdf/Accenture-TechVision-2018-Tech-Trends-Report.pdf#zoom=50.
[4] Valin, J. (2018). Humans still needed: An analysis of skills and tools in public relations. Discussion paper. Retrieved from London: Chartered Institute of Public Relations. https://www.cipr.co.uk/sites/default/files/11497_CIPR_AIinPR_A4_v7.pdf.
[5] Searle, J. (1980). Minds, brains and programs. Behavioral and Brain Sciences, 3, 417-457. doi10.1017/S0140525X00005756.
————————————————————–
We provide precise data, AI Tools and Innovative Solutions!
uMax Data Technology Limited, founded in Hong Kong in 2016, is an innovative technology company driven by artificial intelligence and data. uMax Data offers a variety of SaaS products and solutions, including Social Intelligence Platform (SIP), on-premises AI server, social listening and online alert systems, all-in-one intelligent research and event management platform, AI search engines, social intelligent customer service platforms, AI Agents and chatbots, brand public opinion insights, social media risk management, and multimodal data processing and digital upgrading solutions. These are collectively known as "Social Intelligence." These products not only enhance enterprise productivity and efficiency but also help businesses make more informed decisions.
Want to analyze customer opinions through online discussions? Understand what netizens are thinking? Emailinfo@umaxdata.com immediately to request related case studies and insight reports.