Frontiers in Business, Economics and Management

ISSN: 2766-824X | Vol. 5, No. 3, 2022

Big Data Usage in Marketing Research

Shi Zheng1, Xia Jin1, Wen Zheng2
1 Schoolof Economic and Management, Dalian University for Minorities, Liaoning Province Dalian Development Zone, 116600, China
2 Northeastern
University in Qinhuangdao, Hebei Province, Qinhuangdao, 066000, China
* Corresponding author: Shi Zheng (Email: [email protected])

Abstract: In the marketing field, the use of big data in research can make us understand consumer deeply. In some areas of
market research, big data is already established today. The social media analytics and the use of cookie data to measure internet
coverage are two prominent examples. This essay combs through relevant literatures, discusses the big data uses in the marketing
research and its contribution for decision-making. It presents a revision of main concepts about marketing research, the new
possibilities of use and a reflection about limitations of big data in the marketing research.
Keywords: Big Data, Marketing Research, Marketing Decision Making, User Behavior Track, Data Collection, Behavior
analysis.

can be found, the purpose of research will be achieved. The

1. Introduction statistical data of the statistical department and the data
The key of management lies in decision-making. The key summary of relevant industry associations (alliances), and the
of decision-making lies in information. The more correct the first-hand data obtained by the research team through field
information is, the stronger the relevance is, and the greater research are the main data sources of market opportunities
the probability of success of enterprise decision-making is. At and evaluation.
present, the offline traditional research has the problems of 2.2. Data in Segmentation and Targeting
great difficulty and high cost, while the network access fixed
sample group research and big data research have the In order to succeed, enterprises need to find identifiable,
problems of insufficient sample representation and difficult measurable, substantive, accessible, responsive and feasible
statistical inference. Although a large number of literatures groups. The segmented data comes from both statistical data
are discussing how to make full use of the outstanding and information obtained from continuous research on fixed
advantages of big data, accurately and efficiently reach the samples and information obtained from digital research and
target samples, more comprehensively and profoundly mine monitoring tools. Behavioral segmentation tends to take
user characteristics, gain insight into research content, and advantage of the latter. When the enterprise determines the
complete survey projects with high efficiency and quality; market segment that the competitors do not recognize or the
However, there is still a lack of comprehensive literature to competitors can not provide sufficient services. Through
analyze and summarize the application of big data in market analysis, identifying the market to be subdivided,
marketing research from the perspective of the whole process evaluating how to subdivide, and defining micro
of research. This paper combs the relevant research of segmentation strategies, the segmentation strategies are
scholars and institutions in this field and related fields, determined [1]. Furthermore, the use of score distribution and
systematically refines and summarizes the views of experts advanced analysis technology can realize the calculation of
and scholars on the application of big data in the specific field market dimension and value of life. Consumer behavior and
of marketing research. its beliefs and attitudes can be studied through network
The purpose of this paper is to discuss the application of graphics, which provide relevant information and help
big data in marketing research, put forward the difference enterprises complete the definition of brand and product
between it and the "small data" obtained by traditional positioning.
research, reflect on its limitations and point out the future 2.3. Marketing Research data in Designing and
trend by analyzing its application status and problems in the
marketing research process. For this purpose, the views of Executing Marketing Plan
researchers and experts will be discussed according to Marketing plan is to establish and maintain a feasible
academic publications, and analyzed and faced. Therefore, we adaptive management process between organizational goals,
can infer the conclusion on this topic. skills, resources and its various changing market
opportunities. The formulation and implementation of
2. Marketing Research data in marketing plans will rely on the following four kinds of
Marketing Decision Making research data, (1) product research data. Product development
data may benefit from the history of successful products, the
2.1. Data in Analyzing and Evaluating analysis of process stages, or the query of creative archives
Opportunities [2]. Both online and offline product surveys are necessary.
Retail dive found that 65% of consumers will study the
The success of enterprises in the rapidly changing market products they want online before entering the store. (2) Price
largely benefits from the efforts in consumer research. When survey data. Offline enterprises will also be strongly affected
the opportunities and situations related to the use of products by the use of online price information. Both online and offline

242
price surveys are necessary. Outer box design points out that uneconomical for solving specific decision-making problems.
80% of consumers will query product information, comments This is illustrated by Internet coverage research data based on
and prices through smart phones when shopping in physical cookies or network centric networks. Even though almost the
stores [3]. In April, 2013, a study released by Google Shopper entire population is the same as the census, key information
Marketing Council [4] showed that 54% of consumers would is missing, such as socio demographic statistics [4].
use smartphones to compare prices. (3) Distribution research
data. Information collection activities that affect the selection 3.2. Difference Source
of retail outlets and warehouses, together with the operation The data of market research comes from internal and
information of retailers and wholesalers and their response to external data of the enterprise, as well as field research data.
the marketing policies of manufacturers, are the focus of There are five basic sources of second-hand data: Book and
distribution research. Harvard Business Review research journal publishers, government sources, mass media sources,
shows that 73% of consumers use multiple channels in the industry associations sources, and commercial sources. There
shopping process. Compared with single channel customers, is some overlap between small data and big data. Big data
Omni channel customers' in store consumption will increase expands the sources of primary and secondary data, such as
by 4% and online consumption will increase by 10%. Every mass media, commercial sources, etc. Personal data (data
time they use more channels, customers will pay more money. collected from fidelity program or email), browsing data
(4) Promotional research data. Advertising research and collected through cookies and outsourcing data collected
media research, as well as data on prize sales, coupons, from financial institutions, census and credit cards, UGC data,
sample transactions and research reports on other promotional device data, mobile phone data in log data, web search data,
issues constitute promotional research data. online marketing data, etc. all enrich the existing data forms,
including both text information and image information [5].
2.4. Monitoring Information of Marketing The log information of online marketing activities (i.e.
Performance transaction time, price, volume, etc.) is mainly collected from
Obtaining the monitoring information of brand retail sales relevant online market platforms (take Taobao [16] and
and its market share can analyze the marketing performance Amazon as examples [7]) through the web crawler tool [7].
of enterprises and implement control. In addition, some AIS data in the market forecast is collected and provided by
enterprises will use the data in daily tools such as Google airborne transceivers, ground and satellite base stations [9].
Analytics to evaluate the performance of various marketing Big data of industrial manufacturing is identified as having
channels, including activities, e-mail marketing, paid three sources: structured data, unstructured data and semi-
advertising, social media marketing, content marketing and structured data. Market data, consumption data and sensor
website marketing, evaluate the business contribution of data are classified as structured data; Human interaction data
marketing, extract original data to build dashboards, analyze and social media data are considered as unstructured data;
personal account databases and contact databases, and Web pages and industrial Internet of things data are classified
analyze paid advertising and social media, Use a common as semi-structured data [6].
framework to analyze marketing activities, and analyze
individual marketing activities at any time.
3.3. Difference Design for Deepening Insight
Traditional research includes questionnaire design and big
3. Big Data vs. “Small Data” data research includes algorithm design. This is the soul of
research. Its essence is to deepen insight through data
3.1. Difference Aims: Problem-oriented vs. non validation.
problem-oriented The traditional data research mainly adopts the method of
The data of market research, which we call "small data" in sampling survey, such as the traditional questionnaire survey.
this article, is collected and analyzed problem oriented. In first tier cities such as Beijing and Shanghai, the effective
Whether it's qualitative market visits or quantitative data sample size is more than 300, which is basically considered
research, we must study with problems. We can't do research to represent the whole market. But in the era of big data, in
for the sake of research, otherwise we can only get a lot of theory, all users can be scanned. The increase of sample size
information garbage. The quality of data collected in market will make the results more accurate.
research can be controlled by controlling sampling error and In recent years, China's top domestic entertainment traffic
non sampling error. The relevant variables of measurement has been occupied by small fresh meat. From Weibo topics,
are optimized for specific topics, which can be very extensive. popularity, to the number of fans, to the initiative of the rice
By controlling the sampling error, the representativeness of circle, they firmly occupy the commanding height of the
samples can be guaranteed. industry; This has led to a phenomenon that a large number of
Big data is non problem oriented, and Volker Bosch also female products, such as cosmetics and luxury bags, are
pointed out that big data belongs to passive measurement [5]. endorsed by male celebrities, especially those multinational
They are basically domain data, such as UGC data, device brands. The result is that most of them are not satisfactory.
data, log data [6]. Because information is defined as the lack Since 2022, brands have changed back to female spokesmen.
of uncertainty in statistical analysis, big data is mixed in This is a typical case of lacking algorithm design and being
solving specific problems because it does not reduce brought into the ditch by data; The fans of traffic male stars
uncertainty, and some of them lack important variables. have higher viscosity and are more willing to make lists for
Therefore, when specific research problems are raised, it will their idols, so they can crush female stars in data. However,
be difficult to find big data, whether the scope of data topics the choice of spokesperson is not only a dimension of traffic
and research topics are consistent, and whether the data, but also a consideration of brand and product matching,
terminology and variable classification used by data are relevant associations, public influence and other aspects; If
suitable for research projects, Therefore, big data is we only consider the flow data, there will be no correct

243
research conclusion. The dynamic parameters of behavior trajectory are
First insight, then data; The data reflects the appearance of obtained around consumption behavior, such as user
phenomena and cannot directly get insight, but it can identification problems [15], search information, how to
corroborate and correct the insight to a certain extent. make decisions, purchase frequency, purchase quantity, etc.
The behavior trajectory can abstract the labeled user model.
4. Use of Big Data in Marketing Current research on user behavior trajectory focuses on
Research user web behavior trajectory [19], mobile user behavior
trajectory [20], user trajectory recognition model [21], user
Big data technology makes the information obtained from trajectory recognition method and user trajectory recognition
marketing research more extensive. In the long run, big data system [15][22].
will change marketing research [5]. The current marketing Clickstream is an open source project based on Java, which
research related to big data can be roughly divided into three is mainly used to track the browsing history of users on the
stages: the first stage: the data mainly comes from the OLTP web server. Through the historical data, we can analyze the
data of the main business, including users, commodity path, bottleneck, hotspot connection and so on. The collection
information, transaction data and collection data. The second of user behavior trace information is mainly from the page
stage: with the emergence of personalized recommendation and background, based on the open source clickstream user
system and advertising system, log data has become another behavior collection technology, combined with JavaScript
data source, and some data can be obtained through cookies. script plug-ins, log interception and other methods [23].
The third stage: with the emergence of apps, the behavior For the grasp of product trajectory, it is necessary to master
information of wireless users cannot be obtained through the purchase, sale and inventory of dealers, second batches
traditional logging; Therefore, a number of SDKs developed and terminals, so as to track each product in detail.
by enterprises have emerged, through which wireless Product trajectory refers to the trajectory of products from
information can be directly collected. For example, leaving the factory to circulation, to being purchased and used;
enterprises such as Umeng, talkingdata and Taobao have The most difficult and important thing to master is the
similar interface programs. psychological trajectory, which is also composed of static
The purpose of the market research is to find the trend and parameters and dynamic parameters.
track of the market. Trends of the market include industrial Static parameters refer to users' personality characteristics,
policies, scale, industry growth, profitability, consumption behavior preferences, consumption preferences, values
trends, technological changes, upstream changes, channel tendencies, and dynamic parameters refer to demand
changes, category development, major trends of head brands, motivation, attitude, cognitive characteristics, risk awareness,
etc; Through the inductive analysis of these data, the overall sense of experience, expectation, and satisfaction.
trend judgment is finally obtained. User behavior trajectory The collection of user behavior trajectory information is
will be described in detail later. achieved through JS code insertion, HTTP access interception,
For digital marketing in the era of big data, the first thing java code dynamic implantation and other technologies.
to do is to "clean" and mine the data, refine and analyze it in Therefore, the implementation of user behavior trajectory
depth, and finally form the feasibility report data for decision- requires the cooperation and transformation of the original
making or prediction. Big data is non problem oriented. In system with a certain amount of work [23].
marketing research, it has two main research directions: Data mining tool is an effective method of analyzing and
consumer insight and prediction. As GfK research found, big processing data to access all data sites for multidimensional
data lacks important variables in most marketing research; In analysis and decision support [24].
addition, due to the large number of cases in the big data on UGC data, web data and transaction data of various mobile
consumer behavior, it is still inconclusive to study whether applications provide big data sources for data mining. Now,
the big data can provide the measurement of consumers and various data mining technologies have been applied to extract
groups with such behavior at the same time [6]. user behavior and feature knowledge from UGC data (for text,
4.1. Use of Big Data in Market Research common tokenization tools and stem analysis technology are
used; for photos, gray processing and GLCM method are used.
4.1.1. Analysis of User Behavior and Characteristics
4.1.2. Customer Screening and Precision Marketing
In the era of big data, in theory, all users can be scanned.
Analysis
The increase of sample size will make the results more
Big data analysis enhances customer service and product
accurate. Through the tracking of big data, we can understand
quality for enterprises [25]. Big data analysis is also an
customers in the whole process of customer problem
intangible resource. It is no longer based on experience and
identification, search, evaluation, decision-making, and after-
intuition to guide work, but an ability to develop, explore,
sales [7-14]. According to what users have done in the past,
accumulate, share, and transform information and knowledge
we can infer what they will do in the future, which is the
to guide decision-making [26]. What many entrepreneurs
analysis of user behavior and characteristics, or user behavior
struggle with is that they can distinguish the most valuable
trajectory analysis.
users among the users, friends and fans of the enterprise. They
User behavior trajectory is composed of static and dynamic
can find countless pieces of information through big data, and
parameters. The static parameters in the field of consumer
use some rules to associate and synthesize them, which can
goods are mainly population parameters, age, residence,
help enterprises screen key target users.
gender, income, educational background, family status,
Based on the detailed log information of online marketing
occupation, belief, social class, etc; The static parameters of
(about consumers [27], transaction time [8], products [7],
enterprise users are industry, scale, income, business
transaction price [7], trade volume [8], etc.), we can deduce
characteristics, organizational process, reputation, etc; The
new knowledge about customers' marketing behavior,
platform likes to say "user portrait", which is this part [17].
preferences and interests [8], which seems to be a powerful

244
predictor in relevant predictions. Effective big data can also the probability of something in the future. The biggest
introduce analysis technology to capture other interesting difference between big data and traditional data analysis is to
features, such as the potential aspects of text. Rating analysis make the analysis change from "facing the past that has
is used to estimate personal opinions and the importance of happened" to "facing the future that is about to happen". CPI
each aspect [28]. Another example is semantic based image represents the price fluctuation that has occurred, and big data
retrieval to mine information data hidden in p hoto semantic may help people understand the future price trend and predict
annotation [29]. inflation or economic crisis in advance. The design of text
As an important task of text analysis, text classification has mining technology for CPI prediction [40], the use of Internet
been widely studied, and many methods have emerged, search volume to predict CPI [41], and web crawler
such as text classification method based on hidden document technology to promote the integrated development of big data
body generation model of late Dirichlet allocation (LDA), and CPI survey [42], are all applications of big data in price
text classification method based on bag of word model and trend prediction.
text classification method based on support vector machine In terms of sample time, the application of big data can
(SVM). Research on customer satisfaction through text vector obtain the price and quantity of goods in real time, and
selection of text data (using popular LDA and emotion eliminate statistical errors by dynamically calculating the CPI
analysis tools) has made some progress [30][31]; Some index. In terms of sample range, the application of big data
scholars also extract useful information through information makes the collected samples very similar to the population,
feature extraction of massive features (using PCA and cluster and the range of samples has been significantly expanded.
analysis) [32] [33][34] for user analysis, behavior prediction, After the application of big data, it effectively improves the
customer classification, customer loyalty analysis, etc; speed of CPI weight adjustment, further improves the
Customer behavior in marketing [35] (for example, audio transparency of CPI weight adjustment, and highlights the
data and video data from Netflix, youtube, vine and Facebook) impact of big data on CPI weight [43].
are also used in market forecasting. In data analysis, dealing
with such data types will only slightly increase the complexity. 5. Limitations
In the prediction analysis, AI models (especially support
vector regression (SVR), CNN and long-term and short-term 5.1. Decision-making Is Still the Responsibility
memory (LSTM)) dominate the prediction models using such of Enterprise Decision Makers in The Big
UGC data [36][37][38]. Data Era
4.1.3. Effectiveness of Brand Communication and Big data improves the efficiency of marketing research, but
Analysis of Competitors research is still only responsible for analysis and suggestions,
The effectiveness of brand communication can be found and decision-making is still the responsibility of enterprise
through big data analysis. For example, communication trend decision makers.
analysis, content feature analysis, interactive user analysis, The purpose of the research is to find the trend and
positive and negative emotion classification, word-of-mouth trajectory of the market. It is to assist decision-making, and it
category, analysis, product attribute distribution, etc. can be is no exception in the era of big data. Big data cannot form
carried out. The communication situation of competitors can decisions independently, and there will be many cases of
be mastered through monitoring, and the industry benchmark improper application of big data in the future. Like apple, it
user planning can be referred to. According to the user voice will bear the huge cost of third-party research data support
planning content, the operation effect of microblog matrix can that despises independent systems.
even be evaluated. The automation of decision-making realized by the
Many enterprises want to know about the dynamics of their prediction model [5] also represents risk, because no matter
competitors. Even though it is difficult to obtain the how good the prediction model is, the model is still a binary
information of their competitors, they can analyze it through method to understand the finite theoretical situation. At least
big data monitoring. Collect product review data from various for now, the analytical model will be responsible for the
Internet platforms such as shopping websites, social media implementation of analysis and recommendations, but
and product communities, e.g. Qiu Haibin establishes the co- decision-making is still the responsibility of mankind.
occurrence matrix of target enterprise products and
competitor products from massive data through consumer 5.2. Big Data Is Not Accurate, But Mixed, And
UGC data, with the help of word frequency co-occurrence Needs to Be Used Correctly
technology, carries out visual analysis with the help of social In the era of small samples, data is scarce, so we first pursue
network technology, obtains the specific competitive products the accuracy of data, and then the accuracy of results.
of the target enterprise, and then statistically calculates the However, in the era of big data, data is too complex to be
product sales volume, comment emotional value, product accurate. Some scholars pointed out that in view of the low
network popularity and price of target products and information density of big data, big data is a lean mine, and
competitive products, and establishes the regression model of the output ratio is not necessarily good. In the era of big data,
product sales volume, With the help of TensorFlow neural the hybridity of data becomes inevitable: first, as the amount
network framework developed by Google, a neural network of data becomes larger, the error rate will also increase.
model for predicting product sales is established [39], and the Second, data comes from various systems, which increases
product competitor identification and sales volume prediction the degree of promiscuity. Third, data formats are various, and
are completed. data cleaning is very difficult. In addition, there are data
4.2. Price Trend Forecast shortages and discontinuities due to big data. Therefore, big
data (almost always) is not representative; Big data is (almost
Predictive analysis is the core function of big data. Big data always) flawed; Big data (almost always) lacks important
prediction is based on big data and prediction model to predict variables; These conclusions have almost become an

245
academic consensus [5]. the new manufacturing environment is a challenging problem
It is estimated that only 5% of the data is structured and can [45]. With the further development of data storage and
be stored in traditional databases. If mixing is not accepted, analysis technology, big data-driven analysis is an important
the remaining 95% of unstructured data cannot be used, such driving force to create the main value of manufacturing [46].
as web pages and audio and video resources [5]. By accepting
uncertainty, we open a window to a world that has not been 6. Final Considerations
touched. Big data requires us to change. We must be able to
As discussed above, due to the many characteristics of big
accept confusion and uncertainty. Accuracy seems to have
data, the ultimate problem in the application of big data in
always been the support of our life, but the idea that there is
market research is: who should have the right to collect big
only one answer to each question is untenable. Once we
data, who should have the right to use big data, and who
acknowledge this fact and even support it, we will be one step
should have the right to open big data. As Nunan and
closer to the truth.
Domenico [50] put forward, autonomy through ESOMAR
In the era of small data, we need to use the least data to get
(European Society for public opinion and Market Research)
the most information. The essence of big data is not the
[51] is another way to combat abuse and excessive behavior,
absolute value of the amount of data, but the method of using
And make good use of information. Nunan and Domenico [50]
all available data for analysis instead of random sampling
proposed that the current esomar[51] rules include the right to
analysis. In this sense, big data emphasizes not data, but
be forgotten (which may require deleting the history), the
methodology.
right to expire the data (which may supplement the forgotten
Big data can make up for the deficiencies of traditional
right, and the transaction data may also expire), and the
research, but big data research can't do what traditional
ownership of the social graph (individuals should know the
research can't do. One of the most prominent is the grasp of
collected information). In 2016, a number of manufacturers
users' psychological path. Without the analysis logic of
cooperating with Talkingdata found that the products released
professional research, without the support of data analysis
on Google play were removed from the shelves, and the
technology and huge equipment, the original big data is just a
instructions for the removal said in the email, "in violation of
huge "waste". Therefore, an important aspect of big data
the developer terms" and pointed out that it was caused by the
application is the correct use of information [44].
problems of talkingdata SDK. Talkingdata responded to the
5.3. Big Data Still Lacks Clear Decision SDK, causing it to be removed from the shelf: google play
Guidance reviewed and adjusted. This may be a typical case of privacy
infringement.
Big data implements value orientation rather than decision- In addition, big data must be scientifically evaluated in
making demand orientation, that is, mining in massive data to market research. Passive measurement is the efficiency of big
see what valuable information can be found. In a data in market research. It created what was once a dream,
macroeconomic or social sense, there is no problem at all. that is, economies of scale. Big data can be used like a
However, from the perspective of enterprise micro decision- microscope to see the fuzzy structure, even completely
making, big data with large amount of data, great noise and unrecognizable through traditional market research [5].
low value density may be difficult to meet the applicability However, big data is not omnipotent. It is flawed. There must
requirements of enterprise decision-making for specific data be appropriate personnel in the field of application and
value. research after evaluation. At this level, abuse and excessive
There is no right to speak without investigation, but how behavior should also be avoided. According to gobble (2013)
much say research should have in the final decision-making [2], the McKinsey report identified 190000 qualified
has always been controversial; The general understanding is personnel working in data analysis related positions today. In
that research is to collect information, and insight is to process addition to those directly related to the analysis, don Schulte
information, which can break the essence; Insight without [47] still emphasizes the need for people with "real-life"
research is blind and blind; Research without insight is experience to be able to interpret the information generated
mindless and formalistic. In the information age, it is easier by the algorithm. "If the basic understanding of customers
for enterprises to obtain information and people's behavior does not exist, it is built into the analysis model, and it does
trajectories to be mastered; This makes it more important for not matter how many iterations or how fast the data goes
enterprises to obtain more accurate and effective information through. This output is worthless. Therefore, the construction
and establish market intelligence systems for enterprises of enterprise public information disclosure system [48][49]
through big data. Big data intelligence replaces some of the may be a promising way to promote the application of UGC
previous ways of obtaining information, such as reading data in prediction.
media reports, such as inquiring about various gossip; But big
data cannot replace insight, let alone make decisions References
automatic; Research and insight still perform their respective
duties, make insights into big data information, and use big [1] Cravens, D.W. and Piercy, N.F. (2008) Marketing estratégico.
data information to verify and deepen insight conclusions; 8th Edition, McGraw Hill, São Paulo.
This is the essential relationship between the two and will not [2] Gobble, M.M. (2013) Big Data: The Next Big Thing in
change. The value of big data to enterprise marketing is not Innovation. Research-Technology Management, 56, 64-
only reflected in the information system, but also plays an 67.http://dx.doi.org/10.5437/08956308X5601005
important role in sales management, promotion management, [3] https://www.sohu.com/a/314506582_700450
product and brand management, supply chain management,
[4] Google Shopper Marketing Agency Council (2013) Mobile In-
but the basis is the information system. Store Research: How Is Store Shoppers Are Using Mobile
How to deal with these ultra-high dimensional data, mine Devices, 37.
its potential value, and develop a data flow model suitable for

246
 http://www.marcresearch.com/pdf/Mobile_InStore_Research_ [20] WANG Su-fang. Analysis of mobile user behavior in big data
Study.pdf environment. Information Technology and Informatization,
2020,11:250-252
[5] Volker Bosch. Big Data in Market Research: Why More Data
Does Not Automatically Mean Better Information. GfK MIR / [21] ZHANG WeiLI YangZHANG JiWANG Jian-Yong .A User
/ GfK Research. Vol. 8, No. 2, 2016,56-63 Trajectory Identification Model with Fusion of Spatio-
Temporal Behavior and Social Relation. Chinese Journal of
[6] Chunquan Li , Yaqiong Chen , Yuling Shang.A review of Computers. 2021,44(11):2173-2188.
industrial big data for decision making in intelligent
manufacturing. Engineering Science and Technology,an [22] Wang Yunan. User Behavior Analysis Method Research Based
International Journal.2022,29:1-16 on Trajectory Data. Shenyang Ligong
University,2020.DOI:10.27323/d.cnki.gsgyc.2020.000462.
[7] S. Carta, A. Medda, A. Pili, D.R. Recupero, R. Saia,
Forecasting e-commerce products prices by combining an [23] LinXueyun. Research on User Behavior Trajectory. Journal of
autoregressive integrated moving average (ARIMA) model and Chongqing Industrial and Commercial University(Natural
Google trends data, Future Internet 11 (1) (2019) Science Edition).2015,32(01):78-
5,https://doi .org /10 .3390 /fi11010005. 82.DOI:10.16055/j.issn.1672-058X.2015.0001.019.
[8] J. Zhang, A. Simeone, P. Gu, B. Hong, Product features [24] K. Gandhi, B. Schmidt, A.H. Ng, Towards data mining based
characterization and customers’ preferences prediction based decision support in manufacturing maintenance, in: Proc. CIRP
on purchasing data, CIRP Ann. 67 (1)(2018) 149–152, Conf., 2018, pp. 261–265.
https://doi .org /10 .1016 /j .cirp .2018 .04 .020.
[25] R.H. Hamilton, W.A. Sodeman, The questions we ask:
[9] M. Gao, G. Shi, S. Li, Online prediction of ship behavior with Opportunities and challenges for using big data analytics to
automatic identification system sensor data using bidirectional strategically manage human capital resources, Bus. Horiz. 63
long short-term memory recurrent neural network, Sensors 18 (2020) 85–95, https://doi.org/10.1016/j.bushor.2019.10.001
(12) (2018) 4211, https://doi .org /10 .3390 /s18124211.
[26] J. Lee, J.H. Lee, Constructing efficient regional hazardous
[10] M. Inthachot, V. Boonjing, S. Intakosum, Artificial neural weather prediction models through big data analysis, J. Intell.
network and genetic algorithm hybrid intelligence for Fuzzy Syst. 16 (1) (2016) 1–12,https://doi .org /10 .5391
predicting Thai stock price index trend, Comput. Intell. /ijfis .2016 .16 .1.1.
Neurosci. 15 (2016) 1–8, https://doi .org /10 .1155 /2016
/3045254. [27] K. Fang, Y. Jiang, M. Song, Customer profitability forecasting
using big data analytics: a case study of the insurance industry,
[11] A. Goyal, S. Krishnamurthy, S. Kulkarni, R. Kumar, M. Vartak, Comput. Ind. Eng. 101 (2016)554–564, https://doi .org
M.A. Lanham, A solution to forecast demand using long short- /10 .1016 /j .cie .2016 .09 .011.
term memory recurrent neural networks for time series
forecasting, in: Midwest Decision Sciences Institute [28] H. Wang, Y. Lu, C. Zhai, Latent aspect rating analysis on
Conference, IEEE, 2018, pp. 1–18. review text data: a rating regression approach, in: Proceedings
of the 16th ACM SIGKDD International Conference on
[12] J. Wang, R. Hou, C. Wang, L. Shen, Improved v-support vector Knowledge Discovery and Data Mining, ACM, 2010,pp. 783–
regression model based on variable selection and brain storm 792.
optimization for stock price forecasting, Appl. Soft Comput. J.
49 (2016) 164–178, https://doi .org /10 .1016 [29] K. Song, F. Li, F. Long, J. Wang, Q. Ling, Discriminative deep
/j .asoc .2016 .07.024. feature learning for semantic-based image retrieval, IEEE
Access 6 (2018) 44268–44280, https://doi .org /10 .1109
[13] F. Yoseph, M. Heikkilä, D. Howard, Outliers identification ACCESS .2018 .2862464.
model in point-of-sales data using enhanced normal
distribution method, in: 2019 International Conference on [30] DuChangShun. Research on Key Technologies of Public
Machine Learning and Data Engineering, IEEE, 2019, pp. 72– Opinion Sentiment Analysis for Segmented Fields. Beijing
78. Jiaotong University. 2019.

[14] A. Goyal, S. Krishnamurthy, S. Kulkarni, R. Kumar, M. Vartak, [31] LI Rui;ZHANG Wei-bin. Application of Data Mining
M.A. Lanham, A solution to forecast demand using long short- Technology Based on TF-IDF Algorithm and LDA Topic
term memory recurrent neural networks for time series Model in Power Customer Complaint Text . Techniques of
forecasting, in: Midwest Decision Sciences Automation and Applications,2018,37(11):46-50.
InstituteConference, IEEE, 2018, pp. 1–18. [32] Dengyingfan. Research on user analysis and behavior
[15] Meng Wei;Dong Kai .Research on user behavior track and prediction driven by big data [D]. Beijing University of Posts
early warning system based on big data. Cyberspace and
Security. ,2019,10(12):98-102. Telecommunications,2020.DOI:10.26969/d.cnki.gbydu.2020.
002814.
[16] O. Schaer, N. Kourentzes, R. Fildes, Demand forecasting with
user-generated online information, Int. J. Forecast. 35 (1) (2019) [33] Huangpuhancong, Xiaozhaodi, YuyongZhong. Power
197–212, https://doi .org /10 .1016 /j .ijforecast .2018 .03 .005. consumer value classification and application based on entropy
weight method and improved PCA clustering algorithm[J].
[17] KRIEGEL H. Data Science/Data Mining[J].Digital Welt, Modern Electronics Technique,2017,40(07):183-
2019,3 (1):7-8. 186.DOI:10.16652/j.issn.1004-373x.2017.07.048.
[18] YuSong. Mining Association Rules in the Big Data of Guizhou [34] ZhuBangZhu. Evaluation and Improvement of Customer
Cigarette Brand Loyalty Based on the PCA/DEA Model [J]. Journal of Wuhan
(M),2017)https://kns.cnki.net/KCMS/detail/detail.aspx?dbna University of Technology (information and Management
me=CMFD202001&filename=1017195571.nh Engineering Edition),2008(01):109-113.
[19] Kori S , Zhu Y , Yamaguchi K , et al. Analysis of user's [35] S. Tanuwijaya, A. Alamsyah, M. Ariyanti, Mobile customer
behaviour based on search intentions for information retrieval behaviour predictive analysis for targeting Netflix potential
using search engines[C]// 2015 Conference on Technologies customer, in: 2021 9th International Conference on
and Applications of Artificial Intelligence (TAAI). IEEE, 2015. Information and Communication Technology, ICoICT,IEEE,
2021, pp. 348–352.

247
[36] T. Trzci´ nski, P. Rokita, Predicting popularity of online videos [44] Alexandre Borba Salvador, Ana Akemi Ikeda. Big Data Usage
using support vector regression, IEEE Trans. Multimed. 19 in the Marketing Information System. Journal of Data Analysis
(2017) 2561–2570, https://doi .org /10 .1109 and Information Processing, 2014, 2, 77-85
/TMM .2017.2695439.
[45] S. Kumar, K.K. Mohbey, A review on big data based parallel
[37] W. Liu, Z. Duanmu, Z. Wang, End-to-end blind quality and distributed approaches of pattern mining, J. King. Saud.
assessment of compressed videos using deep neural networks, Univ. Comput. Inf. Sci. doi:10.1016/j.jksuci.2019.09.006.
in: ACM Multimedia, 2018,pp. 546–554.
[46] T. Ahmad, H. Chen, W.A. Shah, Effective bulk energy
[38] H. Dou, W.X. Zhao, Y. Zhao, D. Dong, J.-R. Wen, E.Y. Chang, consumption control and management for power utilities using
Predicting the popularity of online content with knowledge- artificial intelligence techniques under conventional and
enhanced neural networks, in:ACM KDD, 2018. renewable energy resources, Int. J. Electr. Power Energy
Syst.109 (2019) 242–258, https://doi .org /10 .1016
[39] QiuHaiBin. Research on product competitor identification and /j .ijepes .2019 .02 .023.
sales volume prediction based on consumer UGC[D]. Jilin
University,2019. [47] Schultz, D. (2012) Can Big Data Do It All ? Marketing News,
November, 9.
[40] Tang Xiaobin;Dong Manru;Xu Rong. Design and Application
of Text Mining Technology for CPI Prediction Based on Big [48] Ling Tang , Jieyi Li , Hongchuan Du , Ling Li ,∗ , Jun Wu ,
Data [J]. Statistical Research,2021,38(08):146- Shouyang Wang. Big Data in Forecasting Research: A
160.DOI:10.19343/j.cnki.11-1302/c.2021.08.012. Literature Review. Big Data Research.2022,27:1-30
[41] Ouyang Mengqian;Zhou Xianbo;Zhu Junmei. Apply Internet [49] J. Wang, B. Zhang, Quality of environmental information
Search Index to Predict CPI in the Big Data Era——Based on disclosure and enterprise characteristics, Manag. Environ. Qual.
Joint Application of LASSO and Kernel Partial Least Squares An Int. J. 30 (2019) 963–979,https://doi .org /10 .1108 /MEQ -
[J]. Quarterly Journal of Finance,2020,14(02):112-136. 11 -2018 -0194.
[42] Hujianwei,Qiuhaiyan. Web crawler technology promotes the [50] Nunan, D. and Domenico, M.Di. (2013) Market Research and
integrated development of big data and CPI investigation [N]. the Ethics of Big Data Market Research and the Ethics of Big
China Information Daily,2019-09- Data. International Journal of Market Research, 55, 2-13.
12(004).DOI:10.38309/n.cnki.nzgxx.2019.001360.
[51] ESOMAR. http://www.esomar.org/utilities/news-
[43] LiAng. Research on the influence of big data on CPI statistics multimedia/video.php?idvideo=57C. J. Kaufman, Rocky
and method improvement [J]. China Management Mountain Research Lab., Boulder, CO, private communication,
Informationization,2019,22(02):193-194. May 1995.

248