SYSTEM AND METHOD OF

SEGMENTING AND TAGGING ENTITIES BASED ON PROFILE MATCHING USING A MULTI-MEDIA SURVEY

FIELD OF INVENTION

[0001] This invention relates to a computerized system that performs emotional profile clustering, segmenting user groups, and assigning emotional codes to entities based on the clustering result.

BACKGROUND OF INVENTION

[0002] Conventional text -based survey has been practiced for many years in an attempt to understand user preferences. Typically, such survey consists of a question and multiple answers all written in text form. This kind of surveys is effective only in gathering the demographic and factual information from users. Since it requires users to exercise his logical thought process to answer questions, it is not effective in capturing users' emotional reflex. However, it is well observed that humans make snap judgments based on instinctive emotional inclination. As a result, there is a severe limitation in using conventional text-based survey to understand the emotional preferences of users. As a result, marketing campaign based on demographic survey alone may not be very effective.

SUMMARY OF INVENTION

[0003] In the light of the foregoing background, it is an object of the present invention to provide an alternative survey method that can capture the emotional reflex of users and segment those users with similar emotional preferences to the same category. Once done, the individual user can be tagged with an emotional code pin-pointing his emotional category and this code can be used in many commerce applications to enhance their search and match capabilities.

1002.002

[0004] Accordingly, the present invention, in one aspect, provides a method for emotional profiling by first conducting an emotional survey to a user group. This survey comprises at least one survey form; and each form comprises a question and a plurality of multi-media objects for the users to choose from. The multi-media objects can be but not limited to images, photographs, video clips, audio files, or any combination thereof. This survey is intended to capture the emotional reflex of the users. Once a set of survey results are obtained, cluster analysis can be performed to segregate the users into at least one cluster. Each cluster represents a user segment with similar emotional preferences, and an emotional code can be assigned to this cluster.

[0005] In a preferred embodiment, each user is assigned an emotional code of the cluster to which this user belongs. This becomes his personal emotional code and is also referred as his emotional fingerprint or emotional DNA.

[0006] In yet another preferred embodiment, an object can also be assigned to an emotional code. This can be done by firstly assembling a user community; obtaining the personal emotional code of each the user in the user community; adding the number of users in the user community belonging to that personal emotional code; identifying the personal emotional code that has the highest number of users; and assigning the personal emotional code as the object emotional code of that object.

[0007] In a variation of the above method, this invention further comprises the steps of refining the emotional preferences to finer details. This is done by (a) identifying one user segment; (b) performing a second cluster analysis on that user segment to obtain at least one sub-cluster; and (c) assigning a sub-emotional code to the sub-cluster. This procedure can be recursively applied so that sub-sub-emotional code for sub-sub-clusters can be created.

[0008] In a further aspect of the present invention, there is further provided a method of matching the emotional code of one entity against another. An entity is defined here as either a human person or an object. An object can be but not limited to a company, a product, a service or a brand. The personal emotional code of a user and object emotional

1002.002

code of an object can be used to compare against each other to arrive at a matching score. The matching score indicates the emotional similarity of the two entities and this will facilitate many existing search and match applications.

[0009] In another aspect of this invention, a computerized emotional profiling system is provided to implement the aforementioned inventive methods.

[0010] In one embodiment, the system is configured to communicate with the user via a desk-top computer, a portable computer, an information kiosk, a wireless mobile phone device, an interactive TV or an Internet TV.

[001 1] In yet another embodiment of the present invention, the personal emotional code can be stored in a non-volatile storage device under the specific user's possession. This device can be a credit card, a debit card, a smart card, an identity card, a Subscriber Identification Module (SIM) card, or a Universal Subscriber Identification Module (USIM) card.

[0012] There are many advantages to the present invention. One advantage is that it can capture the emotional reflex of a user, something that cannot be done by conventional text-based questionnaires. It is well known in the art that consumers make snap judgment decision in their buying habits. Such decision is largely based on their instinctive emotional preferences rather than a logical, analytical thought process. The emotional profiling methodology therefore provides additional insight for market researchers to understand the behavior of the consumers. Such insights cannot be obtained by conventional text- based surveys.

[0013] Another advantage of the present invention is that when all users and objects are tagged with their respectively emotional code, it allows market researchers to perform targeted marketing in the emotional space. The emotional code can become a universally accepted vocabulary of communication in that space so that commerce can zoom into what the consumer actually likes immediately. Similarly, users can also find like-minded users easily.

1002.002

85

BRIEF DESCRIPTION OF FIGURES

[0014] Fig. 1 depicts the main application software systems of the computerized emotional profiling server according to one example of the present invention.

[0015] Fig. 2 shows the major software modules and databases of the computerized 90 emotional profiling server according to another example of the present invention.

[0016] Fig. 3 is the architecture diagram of the emotional profile cataloguing system according to yet another example of the present invention.

[0017] Fig. 4 is an example of the emotional survey form according to another example of the present invention.

95 [0018] Fig. 5 is a program flow diagram for the media survey module according to another example of the present invention.

[0019] Fig. 6 is a simplified view of a survey document.

[0020] Fig. 7 shows the first emotional profile category.

[0021] Fig. 8 shows the second emotional profile category.

100 [0022] Fig. 9 shows the third emotional profile category.

[0023] Fig. 10 illustrates the segmentation tree structure of emotional profiles and emotional codes.

[0024] Fig. 11 is the architecture diagram of the emotional profile design system

[0025] Fig. 12 is the architecture diagram of the emotional profile identification system.

105 [0026] Fig. 13 is the architecture diagram of the emotional profile matching system.

[0027] Fig. 14 shows the components of the data processing device.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

110 [0028] As used herein and in the claims, "comprising" means including the following elements but not excluding others. Moreover, "entity" includes both a living being such as a user, and an object. The term "object" is used herewith to denote a non-living entity such as but not limited to a company, a corporation, a product, a service, or a brand.

[0029] Referring now to Fig. 1, the first embodiment of the present invention is a

115 computerized emotional profiling server (35) that offers a suite of functions for conducting an emotional profile survey. This computerized emotional profiling server (35) further comprises hardware modules and application software systems to provide the services. The hardware modules will be described later. The software systems comprise (a) an emotional profile cataloguing system (36) that sends out emotional survey forms to users,

120 collects the survey results and clusters the results into emotional categories; (b) an emotional profile design system (37) that helps a designer to design a multi-media message that appeals to the emotional preference of a targeted user community; (c) an emotional profile identification system (38) that categorizes a new entity to the emotional categories found by the emotional profile cataloguing system (36), and (d) an emotional profile

125 matching system (39) that matches the emotional profile of a new entity against existing entities. In one implementation, all these four systems are implemented using a common set of software modules and utilizes a common set of databases to store the relevant information. These common modules and databases are shown in Fig. 2. The media survey module (20) is to send out an emotional survey form to a plurality of users. It does so by

130 first identifying a survey document from the survey document database (105). The survey document contains a plurality of survey forms. Each of the survey form consists of at least one survey question and a plurality of multi-media objects; both of which are drawn from the multi- media presentation database (103). A name-list database (104) is used to supply user names to the media survey module (20). The name-list database (104) may contain a

135 plurality of user communities; each user community comprises a plurality of user names for the media survey module (20) to send the survey form to. In a specific embodiment,

1002.002

the user communities may be a Yahoo group, a MSN community, eBay, Loyalty Marketing Group, Direct Mailing Group, and the likes.

[0030] The media survey module (20) also collects the choices that the user has made in 140 answering the survey questions. Other relevant information such as the speed of answering, the multiple choices that the user selected, is also captured to create the survey result record for this user. The survey result record is then stored onto the survey result database (101) for the cluster analysis module (31) and other modules to access. The cluster analysis module (31) segregates the survey result records into multiple clusters or 145 emotional profile categories. (The term 'cluster' and 'emotional profile categories' are used interchangeably here.) Each cluster represents a segment of users that have similar emotional profile and is assigned an emotional code. All users in that segment are then tagged with that emotional code. The emotional code, as well as other clustering information, is stored as one emotional profile record in the emotional profile database 150 (102). The emotional profile design module (32) is to facilitate a designer to design and progressively refine a multi-media message for a targeted user community. After analyzing the result of a general survey, special survey forms with specific questions can be designed to extract the distinct emotional profile of this targeted user community, and the result is used to compose a new multi-media message.

155 [0031] The emotional profile matching module (40) is to tag a new user with an existing emotional code and the emotional code matching module (43) is to match an emotional code input to the system against existing entries in either the personal emotional code database (106) or object emotional code database (107).

[0032] Fig. 2 also illustrates one embodiment of the invention on the way the 160 computerized emotional profiling server (35) interacts with the user and/or the designer. In a typical implementation, the computerized emotional profiling server (35) is connected to a data communication network. In a preferred embodiment, it is a local area network (26). The local area network (26) connects to a web server (24) and also connects to other computing equipments such as the desk-top computer (28a) and the information kiosk

1002.002

165 (29a) shown in Fig. 2. Similarly, other desk-top computers (28), information kiosk (29) and portable wireless devices (30) can also access the computerized emotional profiling server (35) via the Internet (25) through the web server (24). The portable wireless device (30) can be a mobile phone, a smart -phone, or a PDA-phone that can be connected to the Internet (25) via the mobile operator (27).

170 [0033] It is clear to those skilled in the art that there are many ways to interconnect the computerized emotional profiling server (35) to many different kinds of computing devices that the user or designer use. The teaching shown in Fig. 2 should not be construed as the only way to implement this invention. For example, the computing device can be but not limited to an information kiosk, desk-top computer, lap-top computer, palm-top, PDA,

175 data-tablet, smart-phone, interactive TV, Internet TV or any other devices that can prompt the user or designer a message and solicit a response from them. The data communication network can be but not limited to the Internet (25), a cellular wireless network, a wired local area network (LAN), a wireless LAN, wired or wireless metropolitan area networks, or any combination of the above.

180 [0034] Furthermore, the computerized emotional profiling server (35) may be implemented in a distributed computing environment whereby some of the software modules are installed and run on separate computing platforms. As a specific example, the media survey module (20) can be installed on the desktop computer (28) or the information kiosk (29) across the Internet (25). In this case, the survey is conducted in a

185 remote location and the survey result record is transferred back to the computerized emotional profiling server (35) and is stored in the survey result database (101) herein. Likewise, other software modules can be installed and run on separate computing platforms.

[0035] In one embodiment of the present invention, common modules (20), (31), (32),

190 and (40), and (43) in Fig. 2 are selectively chosen to implement the four application software systems (36), (37), (38) and (39) as shown in Fig. 1. The following sections

1002.002

teach the inventive ideas of each software systems; and how this can be realized by the common modules.

[0036] Fig. 3 illustrates one implementation of the emotional profile cataloguing system

195 (36), which is one of the services that the computerized emotional profiling server (35) provides. This system makes use of the media survey module (20), the cluster analysis module (31), and the five databases (101), (102), (103), (104) and (105) to realize the profile cataloguing operation. The media survey module (20) sends out a survey form (21) to a user and prompts the user to choose one or more answers. The survey form (21)

200 comprises at least one question together with a plurality of multi-media objects. Typically, the question is a textual message. The multi- media object can be but not limited to a textual object, an image, photograph, picture, audio clip, sound clip, movie, or any combination thereof. When the multi-media objects are textual objects, then the survey form is similar to a traditional text-base survey form. In a preferred embodiment, the

205 survey form is displayed on the web browser of the user's computing device. A typical web browser is the Internet Explorer from Microsoft. Fig. 4 shows a typical survey form whereby the multi-media objects are image objects. In yet another preferred embodiment, the image is a mouse-clickable object so that when the user clicks on a particular image, the web-browser detects which image the user has chosen and sends this information back

210 to the media survey module (20) as the user's choice of that survey question.

[0037] The survey document database (105) comprises a plurality of survey documents. Each document is intended for a specific application domain and comprises a set of survey forms. For example, one survey document may be for the leisure domain while another one is for the interior home decoration domain. The survey form may be entirely text-

215 based (i.e. the multi-media objects are text objects). This kind of survey form is to record the factual and demographic information about the users such as their sex, age range, income level and the likes. This is the traditional survey questionnaires. An important aspect of the present invention is not only to gather demographic information from the user but also his emotional preferences. Hence in a typical survey document, the survey

220 forms comprise not only pure text-based forms but also forms with rich multi-media

1002.002

objects such as the one shown in Fig. 4. Research has shown that when pure text-based questionnaires are presented to a user, the user will go through a 'considered thought process' to read up the multiple choice answers before he selects the one that is applicable to him. However, it is well known that feelings drive a vast majority of human behavior 225 and choices, and how people feel in the test environment is closer to how they would feel and act in real-life. This kind of inner feelings can better be captured by user's direct emotional response to visual stimuli. Hence a survey form comprising a plurality of images or visual objects (hereafter also referred to a mood board) can capture the emotional reflex of the user that can not be done by conventional text-based questionnaires.

230 [0038] In a preferred embodiment, the operational flow-chart of the media survey module (20) is depicted in more details in Fig. 5. When this module is invoked, it will first go through the initialization process (71). This includes selecting a user community from the name-list database (104), and a survey document from a survey document database (105). Afterwards, it will choose a new user from the user community in step (72), and then

235 choose a new survey form from the survey document in step (73). Next, it sends the survey form to the user in step (74) and collect user's choices in step (75). The user may select one or more choices. These choices, together with the speed of making the choices are captured by this module. If multiple choices are selected, the sequential order of these choices is also recorded. Since the inventive system is to capture the emotional reflex of

240 the user, the speed, and the ordering of the choices reveal much information about the user's emotional preferences and hence is important information for subsequent analysis.

[0039] Afterwards, it will check if all the survey forms have been sent to this user in step (78). If not, it further tests whether the 'preferential questioning flag' is set in step (80). If not, then it will go back to step (73) to choose a new survey form to send to the user, and 245 the program loop is repeated.

[0040] The 'preferential questioning' is a special feature of this invention. If this flag is set, control is passed to step (76) and the user's choice to the current survey form is analyzed. Depending on the user's choice, a new survey form may be composed that is tailored to

1002.002

the user's specific answer. The new survey form is composed in step (77) and then control 250 is passed to step (74) as shown in Fig. 5. Hence with preferential questioning, survey forms may be generated dynamically depending on user's previous choice and the survey result will better reflect the user's specific emotional reflex.

[0041] In step (78), if all the survey forms have been presented to the user, control is passed to step (79). The survey result, which comprises all the choices made by this user, 255 together with the speed of making these choices and all other relevant information, is stored onto the survey result database (101) as a survey result record. This module will then check if all users in the chosen user community have been processed. If not, control is passed to step (72) and a new user is chosen. Otherwise, the program exits this module.

[0042] After all the survey result records from the user community have been collected, 260 the emotional profile cataloguing system (36) invokes the cluster analysis module (31) to analyze them. In general, the survey result record has a complex data structure in order to store the multi- facet demographic data and emotional preferences of the user. It may be implemented as a high dimensional matrix, a tree structure or an object-oriented data type. In one implementation, it comprises a vector that records the demographic data of the user, 265 and a multi-dimensional matrix that records his emotional preferences. The multidimensional matrix may further comprise the choice vector that registers the choices made by the user, the speed vector to record the time it takes for the user to make that choice(s) and the sequential ordering vector that registers the ordering of choices if the user selects more than one choice for a question. In a simplified implementation, it is a multi- 270 dimensional vector and statistical clustering technique is used to perform clustering analysis on these multi-dimensional vectors. As a typical example, this module can invoke the SPSS statistical package from SPSS Inc. The cluster analysis module (31) produces at least one cluster from the survey results; and also the statistics associated with this cluster.

[0043] To further illustrate how the clustering module performs the emotional profile

275 cataloguing, a specific example is given here. Fig. 6 is a simplified view of the survey document. The first column (50) with single-word labels 'transportation', 'world

1002.002

issues' ... etc is an abbreviated description of the survey questions. For example, the first question, about transportation may be "Which mode of transportation do you like best?". The first row is the images about the modes of transportation for user to select. Here 280 image (51) is the image of a car and image (53) is the image of a train. The first survey form therefore comprises a question about 'transportation' and also the thirteen images from the first row of images in Fig. 6. Likewise, the second survey form comprises a question on world issue such as "Which picture captures best the most important world issue today?" and the thirteen images from the second row of images in Fig. 6.

285 [0044] There are nine survey forms in this survey document illustrated in Fig.6. The survey result record in this case is a multi-dimensional matrix of 3 rows and 117 columns. As there are 9 survey forms and 13 images in each form in this survey document, there are 117 (9x13) selectable choices. Each column corresponds to one choice. The first row of the multi-dimensional matrix stores the choice(s) that the user makes. Notice that the user

290 can select more than one choice in one question. The second row records the time it takes for the user to make the choice(s) and the third row stores the sequential ordering of choices if more than one choice is chosen by the user.

[0045] In operation, the cluster analysis module (31) shown in Fig.2 retrieves all the survey result records from the survey result database (101) and segregates them into

295 clusters or emotional profile categories. As an illustrative example, three emotional profile categories are produced after the clustering analysis, together with a set of cluster statistics associated with each category. Fig. 7 illustrates the first emotional profile category. The majority of users in this category choose the image of the car (51) for 'transportation' and image (52) for the question on 'holiday'. Notice that there is one

300 image chosen in each row; and these images are the most popular choices among the group of users in this category. Likewise, users in the second emotional profile category choose the image of a train (53) for transportation and sunset (54) for holiday as shown in Fig. 8; and users in the third emotional profile category choose images (55) and (56) for the questions on transportation and holiday respectively in Fig. 9. By observing the

305 choices of each group of users, a designer can then assign an emotional code to each

1002.002

category. For example, users of the first category can be labeled as modernist (57), those of the second as traditionalist (58) and the third environmentalist (59). Alternatively, an arbitrary code may be used for each category, e.g. users of the first category can be labeled as category A, those of the second as category B and users in the third labeled as 310 category C. This clustering statistics, together with the emotional code, and all other relevant information related to this emotional profile category are stored as an emotional profile record in the emotional profile database (102).

[0046] After clustering, users in the same emotional profile category can then be assigned to the same emotional code. This code becomes the personal emotional code of this

315 segment of users. As it denotes the emotional preference of the user, it is also referred as his emotional fingerprint. It is advantageous to use an easy-to-remember name or image to denote the emotional code for future references. In the first example given, the names 'traditionalist', 'modernist' and 'environmentalist' are used. This can become the international emotional code; and when this emotional code is adopted worldwide, it can

320 greatly improve the efficiency of the consumer / commerce communication as will be seen by way of examples in subsequent paragraphs. Alternatively, a very long numeric code containing the information of the specific preferences relating to the matrix described above. Many other ways of coding may be devised by a user based

[0047] In another aspect of the present invention, the emotional code of an object such as 325 a company, a service, a product, or a brand, a movie or music is determined. Though these objects are not human subjects that exhibit emotion, it is nonetheless possible to study the users who use, view, purchase, possess, or own the object and arrive at the emotional code of this object. As a typical example, the following method can be used to determine the emotional code of a department store. Firstly, data can be collected for those who 330 patronize this department store. In this example the patrons of the store become the predetermined criteria by which the user community is assembled. The personal emotional codes of these patrons can then be tallied up; and the code that has the highest number of patrons can be used as the emotional code of the department store. This is but just one method of determining the emotional code of an object and is referred as the object

1002.002

335 emotional code. Accordingly, those skilled in the art may devise many other variant schemes to determine the object emotional code and that will generally fall into the scope of this invention. When a user is tagged with his personal emotional code and an object tagged with its object emotional code, many new applications can be developed to make use of this inventive concepts. Both the personal emotional code and the object emotional

340 code can be stored to their respective personal emotional code database (106) and the object emotional code database (107) as shown in Fig. 2.

[0048] In yet another preferred embodiment, the cluster analysis module (31) can be used to obtain sub-emotional categories of a user segment. Recall that the cluster analysis module (31) is used to segregate the plurality of users into at least one user segment, and

345 each segment is assigned to an emotional code. By feeding a user segment to the cluster analysis module (31) again, a more refined emotional sub-categories can be obtained. In essence, the user segment is segregated into smaller sub-segments, each of which can be assigned a more specific emotional code. This is shown in Fig. 10. The original user segment with emotional code 'traditional (T)' (60) can be further segmented to three sub-

350 categories, each of which are tagged with a more specific emotional code - namely, 'Traditional Country (TC)' (61), 'Country Modern (CM)' (62) and 'Classic Antique (CA)' (63). Likewise, each of these sub-segments can be further segregated into sub-sub- segments as shown in Fig. 10. And the corresponding emotional code becomes more and more specific in describing the emotional profiles of that user segment. Clearly, alternative

355 codes, such as but not limited to a string of numerals or bytes may be used to represent them.

[0049] The emotional profile design system (37) comprises the emotional profile design module (32) and those modules and databases that are used in the emotional profile cataloguing system (36) as shown in Fig. 11. The objective of this system is to assist a 360 designer to arrive at a multi-media message (33) that appeals to the emotional preference of a targeted user community. In a specific embodiment, the multi-media message (33) comprises a message that a company wants to convey to the user community and a multimedia object drawn from the multi- media database (103). In a preferred embodiment, the

1002.002

multi- media object is an image, a picture or a photograph. This image will arouse the

365 instinctive emotional reflex of the targeted user so that he is attracted to pay more attention to the message. In operation, the designer first makes use of the clustering result from the emotional profile cataloguing system (36) to understand the emotional preferences of a general user community. He then uses the emotional profile design module (32) to select an emotional profile category from the emotional profile database

370 (102) that he wants to target, and find the multi- media object that is most popular to this user group to compose the multi- media message (33). After presenting this multi-media message (33) to the users in this emotional profile category for a while, a new survey can be conducted on this group with a new survey document that asks more specific emotional preferences questions. After a new set of survey result records are obtained from this user

375 group, the cluster analysis module (31) can be invoked again so that sub-categories of emotional profile can be obtained. The sub-categories of emotional profile are a refinement of the previous ones and can describe the emotional preferences of this user group with better precision. With a more precise emotional profile, the designer can then re-design the multi-media message that appeals to this group. This process can be reiterated so that the

380 designer can better understand the emotional preferences of his targeted user community.

[0050] The emotional profile identification system (38) comprises the media survey module (20) and the emotional profile matching module (40), and accesses the multimedia database (103) and the emotional profile database (102) as shown in Fig. 12. In one implementation, this system first sends survey forms (21) to a new user and collects the

385 survey result record from him. Then it computes matching scores between the survey result record and all the emotional profile records in the emotional profile database (102). If the matching score satisfies a pre-defined criterion, the system will assign the new user to the corresponding emotional profile category and his personal emotional code will be tagged accordingly. The operation flow of the media survey module (20) to collect the

390 survey result is the same as previously discussed and is not repeated here. The emotional profile records, in one specific implementation, comprises an emotional code, the user segment associated with this emotional profile category, and also the cluster statistics computed from the cluster analysis module (40). These emotional profile records are

1002.002

obtained from previously-conducted surveys on the user community. Using the previous

395 example as an illustration, part of the survey result record is the choice vector (63) of 117 elements, each of which is set to '1 ' if chosen by the user; otherwise, it is set to zero. The emotional profile category also has a corresponding vector of similar dimension. The value of each element in this vector is the total number of times this choice is selected by the ensemble user community. The matching score can simply be the inner product (or dot-

400 product) of the two vectors, normalized by the number of choices the new user selected in his survey result record. Alternatively, other methods to compute the matching score can be developed. As an illustration, one can partition the 117 element vector into 9 regions, each of which corresponds to one survey question. Within each region, it further selects the highest value and assigns the corresponding vector-element to one, with the rest reset

405 to zero. This is equivalent to selecting the most popular choice within one question as representative of that question. Afterwards, the inner product is computed and the result is the matching score.

[0051] It is clear to one skilled in the art that many alternative ways to compute the matching score can be devised and they will not be elaborated here. It suffixes to illustrate 410 that a matching score can be computed between the new user's survey result record and the emotional profile record, and this score is then stored as one entry in the matching list (41).

[0052] After all the emotional profile records in the emotional profile database (102) are processed, the matching list (41) is sorted in descending order. In one preferred

415 embodiment, the pre-defined criterion mentioned above is an absolute threshold. In this case, the user is assigned to the corresponding emotional codes whose matching scores are higher than the absolute threshold. In another preferred embodiment, the pre-defined criterion is the top-N entries of the matching list (41), where N is a numerical integer between one and the size of the matching list, and is specified by the designer. As an

420 example, assuming that the highest matching score corresponds to the emotional profile record whose emotional code is 'traditionalist'; and the second highest score corresponds to the emotional code 'environmentalist'. Then if N is 2, the new user is assigned to both a

1002.002

'traditionalist' (top score) and an 'environmentalist' (2 ^nd top score). But if N is 1, he is assigned to the 'traditionalist' only. Hence after going through the survey, each new user 425 is assigned to at least one emotional code. This emotional code becomes his personal emotional code - or Emotional DNA

[0053] The emotional profile matching system (39) is to match the emotional profile of a new entity with an existing entity. It comprises the emotional code matching module (43) and both the personal emotional code database (106) and the object emotional code

430 database (107) as shown in Fig. 13. As mentioned previously, an entity can be a person or an object such as a company, a product, a service, or a brand. This module is capable of (a) matching between two personal emotional codes, (b) matching a personal emotional code against an object emotional code, or (c) matching between two object emotional codes. In operation, the emotional code (64) of a new entity is input to this system. It is used to

435 match with records in either the personal emotional code database (106) or the object emotional database (107) to produce the matching list (41). The emotional code matching module (43) simply compares the emotional code input to the module with those from the databases, and output a list of database entities that matches the input.

[0054] When emotional codes are assigned to all entities, many existing applications can

440 be enhanced and new applications can be enabled - for example, commerce can find consumers, consumers can find consumers and consumers can find commerce. In one specific embodiment, the emotional code is displayed on an Internet search engine homepages and on consumer product and service websites. Consumers use the emotional code as a mean of filtering their web search according to their taste or style preferences.

445 As a specific example, if a consumer is searching for shoes, he can click on an image corresponding to his personal emotional code and then enter 'shoes' on the search bar.

The search engine then finds the shoes that suites the consumer's emotional preferences.

In another example, an Internet auction site can facilitate a seller to specify the emotional code of the potential buyer of his product; and the buyer specifies his emotional code so as

450 to narrow down the search.

1002.002

[0055] In another preferred embodiment, the user's personal emotional code can be stored in a token under his possession. The token can be a credit card, a debit card, a smart card, the SIM (Subscriber Identity Module) card or USIM (Universal Subscriber Identity Module) card of a mobile phone; or any memory devices that can store data persistently. 455 The user can occasionally synchronize the content of his token with information in the computerized emotional profiling server (35) so as to ensure data integrity and consistency.

[0056] When the user is shopping, his token can review his personal emotional code to the shop keeper so that the latter can select products that best match the user's emotional profile. In another embodiment, when the user enters a shopping mall, and he chooses to 460 reveal his personal emotional code to others, he may receive up-to-date promotional offerings that match his personal emotional code from the shops in the mall. Alternatively, he can set up his token with requests for clothing or food, and his token can negotiate with the object emotional codes of merchants in his vicinity and alerts him when he is close by to a shop or restaurant that suits him.

465 [0057] Another application enabled by this invention is to use the emotional profile matching system (39) to find friends with similar or complimentary emotional preferences. This can simply be done by matching the personal emotional code of one user with those in the personal emotional database (106). The system will then report a list of individuals whose emotional preferences are similar to the user.

470 [0058] This invention can also enhance the existing Internet property matching sites so that home buyers can express his or her emotional preferences of the property that he wants to buy. The buyer specifies the emotional code of the property that he is interested in. Likewise, the seller states a similar emotional code for his property, and the emotional profile matching system (39) can then be used to match the buyers with the sellers.

475 Although property matching is used as an example here, it is obvious that the same concept can be extended to other commercial applications such as vehicle purchasing, hotel booking, etc., and any applications that match one entity with another.

1002.002

[0059] In yet another application enabled by this invention, the popular text-based quizzes can be replaced by quizzes in the visual imagery space. As a specific example, images are 480 posted as questions, players also respond by select images in answering.

[0060] These are only some applications that can be enabled or enhanced using the inventive ideas described here. Those skilled in the art can apply these ideas to many other application domains.

485 Hardware and Software Implementation Details

[0061] Fig. 14 shows an exemplary data processing device 800 consisting of both the hardware 801 and software components 802 that can be used to implement the embodiment of the present invention. The hardware components in the present embodiment further comprises a Central Processing Unit (CPU) 810, memory 811, 490 storage 812, and multiple interfaces such as the peripheral interface 813, network interface 814, input interface 815 and output interface 816.

[0062] CPU 810 can be a single microprocessor or multiple processors combined together. Memory 81 1 can include read-only memory, random-access memory or other memory technologies, singly or jointly combined. Storage 812 typically includes persistence storage 495 such as magnetic hard disk, floppy disk, optical storage devices such as CD-ROM, and semiconductor storage devices such as flash memory cards, or other storage technologies, singly or in combination.

[0063] Input interface 815 is the interfacing components that connect the data processing device 800 to data inputting devices such as keyboard, keypad, pen-based device, mouse 500 or other point devices, voice-input apparatus, scanner or other input technologies. Output interface 816 is the interfacing components for the data processing device 800 to send data to outputting devices such as CRT or flat panel display monitor, printer, voice output apparatus, laud speaker or other output technologies. Peripheral interface 813 may typically include the serial or parallel interface and the USB (Universal Serial Bus)

1002.002

505 interfaces, and other interfacing technologies. Network interface 814 enables the data processing device 800 to exchange information with the external data communication network such as the Personal Area Network (PAN), the Local Area Network (LAN), the Wide Area Network (WAN), the Internet, and other data communication network architectures. The network interface 814 can include the Ethernet interface, the Wireless

510 LAN interface device, the Bluetooth interfacing device and other networking devices, singly or in combination.

[0064] Software 802 further includes the operating system 850, and the four application software systems as shown in Fig. 1. Operating system 850 is to manage all the hardware resources, and schedule executing priorities for all tasks and processes so that the four 515 application software systems can all be executed in an orderly manner.

[0065] It should be understood for those skilled in the art that the division between hardware and software is a conceptual division for ease of understanding and is somewhat arbitrary. Moreover, it will be appreciated that peripheral devices in one computer installation may be integrated to the host computer in another. Furthermore, the

520 application software systems may be executed in a distributed computing environment. The software program and its related databases can be stored in a separate file server or database server and are transferred to the local host for execution. The data processing device 800 as shown in Fig. 14 is therefore an exemplary embodiment of how the present invention can be implemented. Those skilled in the art will appreciate that alternative

525 embodiments can be adopted to implement the present invention.

[0066] The preferred embodiments of the present invention are thus fully described. Although the description referred to particular embodiments, it will be clear to one skilled in the art that the present invention may be practiced with variation of these specific details. Hence this invention should not be construed as limited to the embodiments set forth 530 herein.