Marketing and artificial intelligence

Artificial Neural Networks

An artificial neural network is a form of computer program modeled on the brain and nervous system of humans. ^[4] Neural networks are composed of a series of interconnected neurons functioning in unison to achieve certain outcomes. Using “human-like trial and error learning methods”, “neural networks detect patterns existing within a data set is not significant, while emphasizing the data which is most influential”. ^[5]

From a marketing perspective, neural networks are a form of software tool used to assist in decision making. Neural networks are effective in gathering and extracting information from large data sources ^[5] and have the ability to identify the cause and effect within data. ^[6] These neural nets through the process of learning, identify relationships and connections between data bases. Once knowledge has been accumulated, neural networks can be relied upon and applied to a variety of situations. ^[6]

Neural networks help fulfill the role of marketing companies through aiding in market segmentation and measurement of performance while reducing costs and improving accuracy. Due to their learning ability, flexibility, adaptation and knowledge discovery, neural networks offers many advantages over traditional models. ^[7] Neural networks can be used to assist in pattern classification, forecasting and marketing analysis.

Pattern Classification

Classification of customers can be facilitated through the neural network approach allowing companies to make informed marketing decisions. An example of this was employed by Spiegel Inc., a firm dealing in direct-mail operations that used neural networks to improve efficiencies. Using software developed by NeuralWare Inc., Spiegel identified the demographics of having made a purchase. Neural networks where then they are able to identify the key patterns and eventually identify the customers. Understanding this information allowed Speigel to streamline marketing efforts, and reduced costs. ^[8]

Forecasting

Sales forecasting “is the process of future estimating events with the goal of providing good benchmarks for monitoring actual performance and Reducing uncertainty”. ^[9] Artificial technical intelligence-have Emerged to Facilitate the process of forecasting through Increasing accuracy in the areas of demand for products, distribution, employee turnover, performance measurement and inventory control. ^[9] An example of forecasting using neural networks is the Airline Marketing Assistant / Tactician; an application developed by BehabHeuristics which provides guidance for seat allocation through neural networks. This system has been used by Nationalair Canada and USAir. ^[10]

Marketing Analysis

Neural networks provide a useful alternative to traditional statistical models for their reliability, time-saving characteristics and ability to recognize patterns from incomplete or noisy data. ^[6] [11] Examples of marketing analysis systems include the Target Marketing System developed by Churchull Systems for Veratex Corporation. This support system scans a market database to identify customers. ^[10]

When performing marketing analysis, neural networks can assist in the gathering and processing of information from consumer demographics and credit history to the purchase patterns of consumers. ^[12]

Application of Artificial Intelligence to Marketing Decision Making

Is a complex field of decision making which involves a large degree of both judgment and intuition on behalf of the marketer. ^[13] The huge increase in complexity makes the decision makers make the decision making process almost impossible. Marketing decision engine can help distill the noise. The generation of more efficient management has been recognized as a necessity. ^[14] The Application of Artificial intelligence to decision making through a Decision Support Systemhas the ability to help the decision maker. Artificial intelligence techniques are being extended; to provide forecasts; reducing information overload; enabling communication required for collaborative decisions, and allowing for up-to-date information. ^[15]

The Structure of Marketing Decision

Organizations’ strive to satisfy the needs of the customers, paying specific attention to their desires. A consumer-oriented approach requires the production of goods and services that align with these needs. Understanding consumer behavior aids the marketer in making appropriate decisions. Thus, the decision maker is dependent on the decision maker, and the decision environment. ^[14]

Expert System

An Expert System is a software program that combines the knowledge of experts in an attempt to solve problems through expert knowledge and reasoning procedures. Each expert system has the ability to process data, and then through reasoning, transforms it into assessments, judgments and opinions, thus providing advises to specialized problems. ^[16]

MARKEX (Market Expert) is an expert in the field of marketing. These Intelligent decision support systems act as consultants for marketers, supporting the decision maker in different stages, specifically in the new product development process. The software provides a systematic analysis of various methods of forecasting, data analysis and multi-criteria decision making. ^[14]BRANDFRAME is another example of a system developed to assist marketers in the decision-making process. The system supports branding, retail channels, competing brands, targets and budgets. New marketing input is fed into the system where BRANDFRAME analyzes the data. Recommendations are made by the system in view of marketing mix instruments, such as lowering the price or starting a sales campaign.

Artificial Intelligence and Automation Efficiency

Application to Marketing Automation

In terms of marketing, automation uses software to computerize marketing processes that would have otherwise been performed manually. It assists in effectively allowing processes such as customer segmentation, campaign management and products promotion, to be undertaken at a more efficient rate. ^[17] Marketing automation is a key component of Customer Relationship Management (CRM). Companies are using systems that employ data-mining algorithms that analyzes the customer database, giving further insight into the customer. This information can be used to refer to socio-economic characteristics, earlier interactions with the customer, and information about the purchase history of the customer. ^[18]Varinos Systems have been designed to give organizations control over their data. Automation tools allow the system to monitor the performance of campaigns, making adjustments to the campaigns performance tracking. ^[19]

Automation of Distribution

Distribution of products requires companies to access accurate data. Automation processes are able to provide a comprehensive system that improves real-time monitoring and intelligent control. Amazon acquired Kiva Systems , the makers of the warehouse robot for $ 775 million in 2012. Prior to the purchase of the automated system The Kiva robots are able to be satisfied, product replenishment, and more heavy lifting, thus increasing efficiency for the company. ^[20]

Use of Artificial Intelligence to Analyze Social Networks on the Web

A social network is a social arrangement of actors who make up a group, within a network; There can be an array of nodes and nodes that exemplifies common occurrences within a network and common relationships. He (2011), ^[21]describes a social network as, “the study of social entities (people in organization, called actors), and their interactions and relationships. The interactions and relationships can be represented with a network or graph, where each vertex (or node) represents an actor and each link represents a relationship. “At the present time is a growth in virtual social networking with the common emergence of social networks being white replicated online, for example social networking websites Such as Twitter , Facebook andLinkedIn . From a marketing perspective, analysis and simulation of these networks can help to understand consumer behavior and opinion. The use of agent-based social simulation techniques and data / opinion mining to collect social knowledge of networks can help marketers understand their market and segments within it.

Social Computing

Social computing is the branch of technology that can be used by marketers to social networks. ^[22] Social computing provides the platform to create social based software; email: Conditional Random Field (CRFs) technology. conditional Random Field (CRFs) technology. ^[23]

Data Mining

Data mining involves searching the Web for existing information and opinions. “This area of ​​study is called opinion mining or sentiment analysis. It analyzes opinions, appraisals, attitudes, and emotions towards entities, individuals, issues, events, topics, and their attributes “. ^[21]However, it is possible for this information and analysis to be presented to the potential for research bias. Therefore, objective opinion analysis systems are suggested as a solution to this form of automated opinion mining and summarization systems. Marketers using this type of intelligence to make inferences about consumer opinion should be referred to as “spam”, where they are considered to be “spam” or “negative”. ^[21]

Search engines are a common type of information that is sought after by the user. PageRank and HITS are examples of algorithms that search for information via hyperlinks; Google uses PageRank to control its search engine. Hyperlink based intelligence can be used to seek out web communities, which is described as ‘a cluster of densely linked pages representing a group of people with a common interest’. ^[21]

Centrality and prestige are types of measurement used by the group of actors; the terms help to describe the level of influence and actor holds within a social network. Someone who has many ties within a network would be described as a ‘central’ or ‘prestige’ actor. Identifying these nodes within a social network is helpful for marketers to find out about the trendsetters within social networks. ^[21]

References