感性AI List of Products and Services

Words like "kirakira" in "looks sparkling" and "fuwafuwa" in "fluffy texture" symbolically express the state of things through sound. The term that collectively refers to mimetic words and onomatopoeic words is "onomatopoeia." The classification of onomatopoeic and mimetic words varies among researchers and literature, but in Sensory AI, it is defined as above. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Color perception emotions refer to the feelings that arise from the perception of colors, such as feeling "pure" when seeing "white" or feeling "passionate" when seeing "red." While there are individual differences, there are common tendencies that can be summarized as follows: <Warm and Cool Colors> Red hues are perceived as warm, while blue hues are perceived as cool and refreshing. The former is called warm colors, and the latter is called cool colors. <Advancing and Receding Colors> Even if two colors are at the same distance, red hues are felt to be closer, while blue hues are felt to be farther away. The former is called advancing colors, and the latter is called receding colors. *For detailed information on the glossary, please refer to the related links. For more information, feel free to contact us.

Statistical analysis is a systematic methodology for examining the properties of a certain amount of data and inferring the characteristics of a larger dataset (population) from a subset of data (sample). Examples of statistical analysis include principal component analysis, factor analysis, and cluster analysis. *For detailed content of the glossary, please refer to the related links. Feel free to contact us for more information.*

The SD method, short for Semantic Differential method, is a technique used to measure impressions of a subject being evaluated. It is also possible to measure the emotional impressions of products and services. The SD method was devised by psychologists in 1957, placing adjectives with antonyms at both ends of a questionnaire and obtaining numerical data by having subjects evaluate them on a 5-point or 7-point scale. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Principal component analysis is one of the statistical analysis methods. It is a technique that explains the correlation relationships of data with many variables using a smaller number of variables (principal components). By synthesizing or compressing data with many variables into a few principal components while minimizing information loss, one can succinctly understand the overall picture of the data. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.

Factor analysis is a multivariate analytical method used to identify common factors hidden within multivariate data. This method is based on the idea that unobservable latent factors (common factors) influence multiple observed variables. By identifying factors common to many observed variables, factor analysis helps to understand the underlying structure of the data and clarifies the factors behind consumer behavior and other phenomena. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Quantification theory is applied in a wide range of fields, such as market research. This theory refers to methods that convert qualitative data into quantitative elements and perform multivariate analyses similar to multiple regression analysis, principal component analysis, and discriminant analysis. Quantification theory includes Type I quantification (corresponding to multiple regression analysis), Type II quantification (corresponding to discriminant analysis), and Type III quantification (corresponding to principal component analysis or factor analysis). *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Natural language processing is a technology that analyzes everyday languages like Japanese using computers, focusing on natural words rather than artificial languages like programming languages. Natural language can be ambiguous and have different meanings depending on the context, requiring advanced technology for development. Natural language processing can be broadly divided into two processes: language analysis (language understanding) and language generation. Language analysis refers to the process of converting the meaning of a language expression into a form that can be understood by a computer, while language generation refers to the process of creating language expressions from the internal representation of the computer. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Morphological analysis is a part of natural language processing that divides sentences written in natural language into the smallest meaningful units called morphemes, and analyzes their parts of speech and base forms. Natural language refers to languages used in conversation, such as Japanese and English, which are not programming languages. For example, the English word "cats" is composed of two morphemes: "cat" and "s," where "cat" means cat and "s" indicates plurality. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

There are basically two main approaches to information filtering. <Content-based filtering> One is content-based filtering, which involves extracting features from the target content and modeling preference information based on evaluations of that content. *For detailed information on the glossary, please refer to the related links. For more information, feel free to contact us.*

Annotation refers to the process of adding information such as tags or metadata to data. This allows for a clear understanding of what each data element represents across various data formats, including text, images, audio, and video. In AI development, annotation is extremely important. Annotated data is referred to as training data and is used for the machine learning of AI models. *For more details on the glossary, you can view it through the related links. Please feel free to contact us for more information.*

Genetic algorithms were devised in 1975, and their foundation is based on Darwin's theory of evolution. Among the theories of evolution, the concept of natural selection is particularly important. According to this theory, even within the same species, individual organisms possess different traits, and those individuals with advantageous traits are successful in adapting to their environment and can leave many descendants. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

AI (artificial intelligence) is understood as a concept where computer programs perform intellectual activities similar to humans, but its definition is not generally established. The term was first used in 1956, and in recent years, related terms such as machine learning and deep learning have also become commonly used. *For detailed content of the glossary, you can view it through the related links. Please feel free to contact us for more information.*

Deep learning is a form of machine learning that involves machines learning from large amounts of data to extract features from that data. Deep learning differs from traditional machine learning in that, in traditional machine learning, the manual extraction of features is the initial step, and models are built using those features. However, in deep learning, features are automatically extracted by the machine. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Machine learning is a technology that automatically analyzes data and learns rules and patterns, enabling predictions and judgments when faced with new similar challenges. Deep learning is a type of machine learning where the model automatically learns useful features without human intervention, enhancing its accuracy. There are three types of machine learning: (1) supervised learning, (2) unsupervised learning, and (3) reinforcement learning. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Evaluation AI is an AI that calculates evaluation values for objects based on a large amount of learned data. For example, in AI-driven personnel evaluation systems, it is utilized for talent management by visualizing employees' motivation and health status, while systems that visualize package impressions help streamline consumer research, contributing to improved operational efficiency. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Generative AI, also known as "Generative AI," is a type of AI. Generative AI creates various things such as text, images, and music from the data it has learned. It is expected to serve as a tool to support human work by enabling the creation of new ideas. *For detailed content of the glossary, you can view it through the related links. Please feel free to contact us for more information.*

Personal surveys are one of the survey methods used for data collection. There are various types, and a widely used method today is the internet survey, where respondents select options on a computer or smartphone screen. Other methods include in-person interviews where surveyors visit respondents directly, telephone surveys where surveyors call respondents, and drop-off surveys where questionnaires are delivered to respondents and left for a certain period for them to fill out. *For detailed content of the glossary, you can view it through the related links. For more information, please feel free to contact us.*

A venue survey is one of the survey methods used for data collection. There are several types, including methods that gather participants at a venue for a simultaneous survey, methods that conduct individual interviews, and methods that form small groups for discussions while taking notes. *For detailed information on the glossary, please refer to the related links. For more information, feel free to contact us.*

A concept test is a research method used to evaluate how well the concept or idea of a new product or service will be received by customers before it is introduced to the market. Below are the key points of a concept test. 【Purpose of Concept Testing】 Concept testing is conducted to identify which idea among multiple proposals for a new product or service is most suitable for customers and to assess its likelihood of acceptance. This is utilized in various situations, including new product development, rebranding, advertising campaigns, and launching new businesses. *For detailed information on the glossary, please refer to the related links. For more information, feel free to contact us.*

"Consumer Insight" refers to the deep-seated feelings and motivations that consumers are not consciously aware of. The term "insight," from which it derives, means "perception" or "the ability to see the essence," but consumers are often only aware of a small part of their behavior and tend to make product choices based mainly on a vague sense of "just because." *For more detailed information on the glossary, please refer to the related links. Feel free to contact us for more details.*

A/B testing is a widely used method in web marketing aimed at improving the conversion rate (CV rate). In this method, different designs, catchphrases, and other variables are prepared as multiple variations, which are then randomly presented to users. Afterward, the effects of each variation are compared to identify the elements that are expected to be successful. *For detailed information on the glossary, please refer to the related links. For more information, feel free to contact us.*

DAGMAR is a model for evaluating the effectiveness of advertising and capturing the gradual transformation of consumer psychology. Originally proposed at the request of the American National Advertisers Association (ANA), it was devised by management consultant Colley. As part of the advertising effectiveness hierarchy model, DAGMAR expresses the transformation of consumer psychology through five stages: unawareness, awareness, comprehension, conviction, and purchase. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

AIDMA is an acronym that represents the stages of the purchasing process and is widely used in marketing. Each process consists of five stages: Attention, Interest, Desire, Memory, and Action. Understanding the customer’s behavioral process helps in developing effective marketing strategies. *For detailed information on the glossary, please refer to the related links. Feel free to contact us for more information.*

AISAS refers to a model that represents the customer purchasing process, consisting of the initials of "Attention," "Interest," "Search," "Action," and "Share." It describes the steps where a customer pays attention to a product or service, develops interest, searches for information, ultimately takes action (such as making a purchase), and then shares information afterward. *For detailed information on the glossary, please refer to the related links. Feel free to contact us for more details.*

5A refers to a new concept of the customer journey proposed by Philip Kotler in "Marketing 4.0." Each A represents (1) Aware, (2) Appeal, (3) Ask, (4) Act, and (5) Advocate. *For detailed information on the glossary, please refer to the related links. For more information, feel free to contact us.*

Emotional marketing refers to a marketing approach that emphasizes sensory value, such as beauty, emotion, and empathy, which can be felt through the five senses, in addition to the traditional values of functionality and quality in products. In today's world, where commoditization is accelerating, it has become necessary to provide values that consumers require beyond just functionality and quality. Emotional marketing, through sensory value, can enhance brand strength and create a competitive advantage. *For detailed information on the glossary, please refer to the related links. Feel free to contact us for more information.*

The marketing funnel is a diagrammatic representation of the process that customers go through from becoming aware of a product or service to making a purchase, divided into stages. The term "funnel" refers to a shape that illustrates how customers who become aware of a product or service are filtered down, resulting in a gradually decreasing number of individuals who ultimately make a purchase. Generally, the process progresses in the following order: awareness → interest → comparison and consideration → purchase. *For detailed information on the glossary, please refer to the related links. For more information, feel free to contact us.*

Experience value marketing is a marketing approach that emphasizes "experiential consumption" over "material consumption" in modern times. It focuses not only on the features and benefits of products but also on building sensory and emotional connections. To enrich consumers' lives, marketing strategies are being developed that focus on the experiences and values provided by products and services. These strategies include approaches to the five senses, appeals to emotions, stimulation of intellectual curiosity and imagination, value provision through experiences and communication, and a focus on a sense of belonging to a group. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

Digital marketing refers to the methods of conducting sales activities for one's own products or services based on data obtained through digital channels. The term "channel" refers to the sales pathways. Digital channels include, for example, web advertising, social media, and big data accumulated in physical stores. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*

The laddering method is a technique used to investigate the underlying reasons why consumers choose products and the latent values they hold. In research using the laddering method, multiple questions are asked of consumers to delve deeper into their thoughts. The fundamental question of the laddering method is "Why is that important to you?" By repeatedly asking this question, the underlying reasons and values are revealed. *For detailed content of the glossary, please refer to the related links. For more information, feel free to contact us.*