TWM650999U - Intelligent labeling system - Google Patents
Intelligent labeling system Download PDFInfo
- Publication number
- TWM650999U TWM650999U TW112207050U TW112207050U TWM650999U TW M650999 U TWM650999 U TW M650999U TW 112207050 U TW112207050 U TW 112207050U TW 112207050 U TW112207050 U TW 112207050U TW M650999 U TWM650999 U TW M650999U
- Authority
- TW
- Taiwan
- Prior art keywords
- customer
- intelligent
- information
- referral
- text
- Prior art date
Links
- 238000002372 labelling Methods 0.000 title claims abstract description 22
- 230000003993 interaction Effects 0.000 claims abstract description 37
- 230000008451 emotion Effects 0.000 claims abstract description 12
- 238000003058 natural language processing Methods 0.000 claims description 14
- 230000003068 static effect Effects 0.000 claims description 6
- 238000000034 method Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 abstract description 5
- 230000006399 behavior Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000002452 interceptive effect Effects 0.000 description 4
- 230000003542 behavioural effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012351 Integrated analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
一種智能貼標系統，其中的智能客服裝置透過第一客戶端裝置與客戶互動並蒐集一互動記錄，且分析該互動記錄以識別客戶對特定文字的情感或意圖而產生第一客戶意向資訊，其中的智能外撥裝置撥打語音電話給第二客戶端裝置以與客戶進行語音互動並蒐集一語音互動記錄，且分析該語音互動記錄以識別客戶對特定文字的情感或意圖而產生第二客戶意向資訊；其中的轉介裝置將第一客戶意向資訊、第二客戶意向資訊及客戶的結構化資訊提供給其中的標籤裝置；標籤裝置對第一客戶意向資訊、第二客戶意向資訊及結構化資訊進行綜合分析，以判定客戶對特定文字的喜好程度並給予相對應的標籤，並將與客戶對應的標籤提供給轉介裝置。An intelligent labeling system in which an intelligent customer service device interacts with a customer through a first client device and collects an interaction record, and analyzes the interaction record to identify the customer's emotion or intention for specific text to generate first customer intention information, wherein The smart outbound device makes a voice call to the second client device to perform voice interaction with the customer and collects a voice interaction record, and analyzes the voice interaction record to identify the customer's emotion or intention for the specific text to generate the second customer intention information ; The referral device provides the first customer intention information, the second customer intention information and the customer's structured information to the label device; the label device performs the processing on the first customer intention information, the second customer intention information and the structured information. Comprehensive analysis is used to determine the customer's preference for specific text and to assign corresponding tags, and to provide the tag corresponding to the customer to the referral device.
Description
本新型是有關於一種標註代表客戶行為特徵的標籤系統，特別是指一種挖掘客戶互動紀錄中的資訊以分析客戶需求的智能貼標系統。 The invention relates to a labeling system that represents customer behavior characteristics, and in particular, to an intelligent labeling system that mines information in customer interaction records to analyze customer needs.
現行的標籤系統大都針對系統預先處理過並儲存的客戶的靜態資訊，例如會員編號、購買日期、購買品項、購買金額...等具有固定欄位和固定格式的結構化資訊，分析客戶的行為與特徵(例如產品偏好、通路偏好、生活喜好、運動喜好、購物特徵...等)，並根據分析結果給予客戶相對應的標籤，讓業務人員可以根據客戶的標籤，對客戶提供個人化的行銷或銷售活動。 Most of the current labeling systems focus on the static information of customers that has been pre-processed and stored by the system, such as membership number, purchase date, purchased items, purchase amount... and other structured information with fixed fields and fixed formats, to analyze the customer's Behaviors and characteristics (such as product preferences, access preferences, life preferences, sports preferences, shopping characteristics, etc.), and give customers corresponding labels based on the analysis results, so that business personnel can provide personalized services to customers based on their labels. marketing or sales activities.
但是客戶的靜態資訊的分析往往跟不上客戶的動態資訊，因為客戶對於行銷或銷售活動的意向(意願)或意圖往往被埋藏在例如客戶與線上客服互動的對話內容、文字內容或電子郵件等動態資訊(或稱非結構化資訊)中，但這些動態資訊經常被忽略；因此，若能將這些動態資訊(非結構化資訊)與客戶現有的靜態資訊(結構化資訊)做即時整合性分析，從中獲取客戶的特徵並加以標註(標 籤)，將有助於提升對於客戶意向的掌握度。 However, the analysis of customers' static information often cannot keep up with the customer's dynamic information, because customers' intentions (willingness) or intentions for marketing or sales activities are often buried in, for example, the conversation content, text content or emails between customers and online customer service. Dynamic information (or unstructured information), but this dynamic information is often ignored; therefore, if these dynamic information (unstructured information) can be analyzed in real time with the customer's existing static information (structured information) , obtain customer characteristics and label them (label Signing) will help improve the grasp of customer intentions.
因此，本新型之目的，即在提供一種智能貼標系統，其對客戶的非結構化資訊與結構化資訊進行整合性分析，以即時且更精確地獲取客戶的特徵並加以標籤，而有助於提升對於客戶意向的掌握度。 Therefore, the purpose of this new model is to provide an intelligent labeling system that conducts an integrated analysis of the customer's unstructured information and structured information to obtain the customer's characteristics in real time and more accurately and label them, thereby helping To improve the understanding of customer intentions.
於是，本新型一種智能貼標系統，包括一轉介裝置、一與該轉介裝置電耦接的標籤裝置，一與該轉介裝置電耦接的智能客服裝置及一與該轉介裝置電耦接的智能外撥裝置。 Therefore, the new intelligent labeling system of the present invention includes a referral device, a label device electrically coupled to the referral device, an intelligent customer service device electrically coupled to the referral device, and a labeling device electrically coupled to the referral device. Coupled intelligent outbound dialing device.
該智能客服裝置能與一客戶的一第一客戶端裝置通訊，以透過該第一客戶端裝置與該客戶互動並蒐集一互動記錄，且利用一自然語言處理模型分析該互動記錄，以識別該客戶對一特定文字的情感或意圖而產生一第一客戶意向資訊，並提供該第一客戶意向資訊給該轉介裝置。 The intelligent customer service device can communicate with a first client device of a customer to interact with the customer through the first client device and collect an interaction record, and use a natural language processing model to analyze the interaction record to identify the The customer's emotion or intention for a specific text generates a first customer intention information, and the first customer intention information is provided to the referral device.
該智能外撥裝置主動撥打一語音電話給該客戶的一第二客戶端裝置，以透過該第二客戶端裝置與該客戶進行語音互動並蒐集一語音互動記錄，且利用另一自然語言處理模型分析該語音互動記錄，以識別該客戶對該特定文字的情感或意圖而產生一第二客戶意向資訊，並提供該第二客戶意向資訊給該轉介裝置。 The smart outbound dialing device actively makes a voice call to a second client device of the customer to conduct voice interaction with the customer through the second client device and collect a voice interaction record, and uses another natural language processing model Analyze the voice interaction record to identify the customer's emotion or intention for the specific text to generate second customer intention information, and provide the second customer intention information to the referral device.
該轉介裝置將該第一客戶意向資訊、該第二客戶意向資 訊及該客戶的一結構化資訊提供給該標籤裝置。 The referral device combines the first customer intention information, the second customer intention information Information and a structured information of the customer are provided to the tag device.
該標籤裝置對該第一客戶意向資訊、該第二客戶意向資訊及該結構化資訊進行綜合分析，以判定該客戶對該特定文字的喜好程度並給予相對應的一標籤，並將與該客戶對應的該標籤提供給該轉介裝置。 The label device comprehensively analyzes the first customer intention information, the second customer intention information and the structured information to determine the customer's preference for the specific text and gives a corresponding label, and will communicate with the customer The corresponding tag is provided to the referral device.
在本新型的一些實施態樣中，該智能客服裝置蒐集的該互動記錄包括語音、文字、電子郵件內容至少其中之一，且該智能客服裝置對該互動記錄進行預處理並轉換成一第一文本，再利用該自然語言處理模型分析該互動記錄的該第一文本。 In some implementations of the present invention, the interaction record collected by the intelligent customer service device includes at least one of voice, text, and email content, and the intelligent customer service device preprocesses the interaction record and converts it into a first text , and then use the natural language processing model to analyze the first text of the interaction record.
在本新型的一些實施態樣中，該智能外撥裝置會先將蒐集的該語音互動記錄進行預處理並轉換成一第二文本，再利用該另一自然語言處理模型分析該語音互動記錄的該第二文本。 In some implementations of the present invention, the smart outbound dialing device will first preprocess and convert the collected voice interaction record into a second text, and then use another natural language processing model to analyze the voice interaction record. Second text.
在本新型的一些實施態樣中，該轉介裝置還提供一客戶名單給該智能外撥裝置，且該智能外撥裝置是根據該客戶名單撥打該語音電話給該客戶的該第二客戶端裝置。 In some implementations of the present invention, the referral device also provides a customer list to the intelligent outbound dialing device, and the intelligent outbound dialing device makes the voice call to the second client of the customer based on the customer list. device.
在本新型的一些實施態樣中，該結構化資訊是具有固定欄位和固定格式的靜態資訊。 In some implementations of the present invention, the structured information is static information with fixed fields and fixed formats.
本新型之功效在於：該標籤裝置可以從客戶之不斷更新的互動資訊中即時地分析並挖掘出被埋沒在互動資訊中之客戶對於各種特定文字的想法、意圖或偏好，而能針對客戶的行為特徵更 準確且更即時地更新該客戶的標籤，而有助於提升對於客戶意向的掌握度。 The function of this new model is that the label device can instantly analyze and dig out the customers' thoughts, intentions or preferences for various specific words that are buried in the interactive information from the customers' constantly updated interactive information, and can target the customer's behavior Feature updates Accurately and more immediately update that customer's tags, helping to improve visibility into customer intent.
100:智能貼標系統 100:Intelligent labeling system
1:轉介裝置 1: Referral device
2:標籤裝置 2: Label device
3:智能客服裝置 3: Intelligent customer service device
31:智能客服機器人 31: Intelligent customer service robot
32:自然語言處理模型 32:Natural language processing model
4:智能外撥裝置 4: Intelligent outbound dialing device
41:智能外撥機器人 41: Intelligent outbound dialing robot
42:自然語言處理模型 42:Natural language processing model
5:第一客戶端裝置 5: First client device
6:第二客戶端裝置 6: Second client device
本新型之其他的特徵及功效，將於參照圖式的實施方式中清楚地顯示，其中：圖1是本新型智能貼標系統的一實施例的主要組成裝置方塊示意圖。 Other features and functions of the present invention will be clearly shown in the embodiments with reference to the drawings, in which: Figure 1 is a block diagram of the main components of an embodiment of the smart labeling system of the present invention.
在本新型被詳細描述之前，應當注意在以下的說明內容中，類似的元件是以相同的編號來表示。 Before the present invention is described in detail, it should be noted that similar elements are represented by the same numbers in the following description.
參閱圖1所示，是本新型智能貼標系統的一實施例，該智能貼標系統100主要包括一轉介裝置1、一標籤裝置2、一智能客服裝置3及一智能外撥裝置4。
Refer to Figure 1 , which is an embodiment of the new intelligent labeling system. The
該轉介裝置1(或稱中介裝置或中介平台)是一電腦裝置，其與該標籤裝置2、該智能客服裝置3及該智能外撥裝置4直接電連接或透過網路電耦接，以轉送來自該智能客服裝置3及該智能外撥裝置4的資訊或資料給該標籤裝置2，並接收來自該標籤裝置2的資訊或資料。
The referral device 1 (also called an intermediary device or an intermediary platform) is a computer device that is directly electrically connected to the tag device 2, the intelligent
該標籤裝置2是由一電腦裝置構成的一標籤系統，其用 以從蒐集來的客戶的互動記錄和客戶的既有資訊中找出客戶的特徵並給予相對應的標籤，以標註客戶的行為特徵，使客戶的標籤成為業者跟客戶互動或行銷的基礎。 The labeling device 2 is a labeling system composed of a computer device, which is used to The customer's characteristics are found from the collected customer interaction records and existing customer information, and corresponding labels are given to mark the customer's behavioral characteristics, so that the customer's label becomes the basis for the interaction or marketing between the business and the customer.
該智能客服裝置3是做為一伺服器的電腦裝置，其能與一客戶的一第一客戶端裝置5通訊，並應用一智能客服機器人31透過該第一客戶端裝置5與該客戶互動並自動回答或答覆該客戶的詢問或提問。
The intelligent
該第一客戶端裝置5可以是例如該客戶的室內電話、行動電話、電腦等，且該客戶可藉由撥打客服電話、線上語音或文字諮詢或傳送電子郵件等方式向該智能客服裝置3的該智能客服機器人31提問；同時，在該客戶與該智能客服機器人31互動的過程中，該智能客服裝置3蒐集該客戶與該智能客服機器人31的一互動記錄。
The
因此，該互動記錄可能包括語音、文字或電子郵件內容等至少其中之一。且該智能客服裝置3對該互動記錄進行預處理(例如語音轉文字、從非結構化文字資料中取出有意義的資料等)並轉換成一第一文本，再利用經過訓練的一自然語言處理模型32分析該互動記錄的該第一文本，以識別該客戶對該第一文本中出現的一特定文字(例如該客戶在詢問過程中反覆提到的一產品名稱或一活動名稱)的情感或意圖(意向)而產生一第一客戶意向資訊，並提
供該第一客戶意向資訊給該轉介裝置1；其中該第一客戶意向資訊包含該客戶對該特定文字的情感或意圖，例如若該第一文本是關於該客戶詢問該特定文字”ETF”，且該客戶對該特定文字”ETF”的情感或意圖例如是表現出想了解、有意願購買、考慮購買等。
Therefore, the interaction record may include at least one of voice, text or email content. And the intelligent
該智能外撥裝置4是做為一伺服器的電腦裝置，且其應用一智能外撥機器人41根據一客戶名單主動撥打一語音電話給該客戶的一第二客戶端裝置6，例如客戶的一行動電話，以透過該第二客戶端裝置6與該客戶進行語音互動，例如對該客戶進行針對該特定文字(比如一產品名稱或一活動名稱)的問卷調查或行銷推廣活動等，並蒐集該客戶在答覆過程中與該智能外撥機器人41的一語音互動記錄。
The intelligent
然後，該智能外撥裝置4會先將蒐集的該語音互動記錄進行預處理(例如語音轉文字、從非結構化文字資料中取出有意義的資料等)並轉換成一第二文本，再利用經過訓練的另一自然語言處理模型42分析該語音互動記錄的該第二文本，以識別該客戶對於該第二文本中出現的該特定文字的情感或意圖而產生一第二客戶意向資訊，並提供該第二客戶意向資訊給該轉介裝置1。
Then, the intelligent
該第二客戶意向資訊包含該客戶對該特定文字的情感或意圖，例如若該第二文本是有關調查該客戶對該特定文字”ETF”的看法或了解的意願，而該客戶對該特定文字”ETF”表達正面、肯 定、有意願、負面、否定或拒絕等情感或意圖。 The second customer intention information includes the customer's emotion or intention about the specific word. For example, if the second text is about investigating the customer's views or willingness to understand the specific word "ETF", and the customer's intention about the specific word "ETF", "ETF" expresses positivity and willingness Emotions or intentions such as determination, intention, negativity, negation or rejection.
值得一提的是，該客戶名單是由該轉介裝置1提供給該智能外撥裝置4。
It is worth mentioning that the customer list is provided to the intelligent
該轉介裝置1於收到該智能客服裝置3傳來的該第一客戶意向資訊和該智能外撥裝置4傳來的該第二客戶意向資訊後，該轉介裝置1將該第一客戶意向資訊、該第二客戶意向資訊和該客戶的一結構化資訊提供給該標籤裝置2；該結構化資訊可以是預存在該轉介裝置1中或者由該轉介裝置1向一客戶資料庫裝置(圖未示)取得，且該結構化資訊是客戶的例如會員編號、購買日期、購買品項、購買金額...等具有固定欄位和固定格式的靜態資訊。
After receiving the first customer intention information from the intelligent
該標籤裝置2對該第一客戶意向資訊、該第二客戶意向資訊及該結構化資訊進行即時性綜合分析，例如從該第一客戶意向資訊、該第二客戶意向資訊及該結構化資訊中分析客戶對於該特定文字的行為與特徵，以判定該客戶對該特定文字的喜好程度並給予相對應的一標籤，例如該標籤裝置2分析發現該客戶在一預設的一段時間(例如一個月或一季)對該特定文字比較感興趣且對該特定文字進行了正面的詢問、了解或答覆，而且該結構化資訊亦顯示該客戶曾購買(或參加)與該特定文字相關的產品(或活動)，該標籤裝置2即判定該客戶對該特定文字有高度喜好(或高意向、高意願)，即給予相對應的一標籤，該標籤例如以代表高度喜好的一種顏色為 底色並標註該特定文字(比如ETF)。然後，該標籤裝置2將與該客戶對應的該標籤提供給該轉介裝置1。 The tag device 2 conducts a real-time comprehensive analysis of the first customer intention information, the second customer intention information and the structured information, for example, from the first customer intention information, the second customer intention information and the structured information. Analyze the customer's behavior and characteristics of the specific text to determine the customer's preference for the specific text and give a corresponding label. For example, the label device 2 analyzes and finds that the customer is within a preset period of time (such as one month). or quarter) are more interested in the specific text and have made positive inquiries, understanding or responses to the specific text, and the structured information also shows that the customer has purchased (or participated in) products (or activities) related to the specific text ), the label device 2 determines that the customer has a high degree of preference (or high intention, high willingness) for the specific text, that is, gives a corresponding label. The label is, for example, a color representing a high degree of preference. background color and mark the specific text (such as ETF). Then, the tag device 2 provides the tag corresponding to the customer to the referral device 1 .
藉此，根據上述分析流程，該標籤裝置2可以從客戶之不斷更新的互動資訊(動態資訊)中即時地分析並挖掘出該客戶之被埋沒在互動資訊中之對於各種特定文字的想法、意圖或偏好，而能針對客戶的行為特徵更準確且更即時地更新該客戶的標籤，而有助於提升對於客戶意向的掌握度，讓業務單位能夠根據客戶的標籤對客戶進行更精準及個人化的行銷，確實達成本新型的功效與目的。 Thereby, according to the above analysis process, the labeling device 2 can instantly analyze and dig out the customer's thoughts and intentions for various specific words that are buried in the interactive information from the customer's constantly updated interactive information (dynamic information). or preferences, so that the customer's label can be updated more accurately and more immediately based on the customer's behavioral characteristics, which helps to improve the grasp of customer intentions and allows business units to more accurately and personalize customers based on their labels. The marketing has indeed achieved the efficacy and purpose of this new model.
惟以上所述者，僅為本新型之實施例而已，當不能以此限定本新型實施之範圍，凡是依本新型申請專利範圍及專利說明書內容所作之簡單的等效變化與修飾，皆仍屬本新型專利涵蓋之範圍內。 However, the above are only examples of the present invention, and should not be used to limit the scope of the present invention. All simple equivalent changes and modifications made based on the patent scope of the present invention and the content of the patent specification are still within the scope of the present invention. Within the scope covered by this new patent.
100:智能貼標系統 100:Intelligent labeling system
1:轉介裝置 1: Referral device
2:標籤裝置 2: Label device
3:智能客服裝置 3: Intelligent customer service device
31:智能客服機器人 31: Intelligent customer service robot
32:自然語言處理模型 32:Natural language processing model
4:智能外撥裝置 4: Intelligent outbound dialing device
41:智能外撥機器人 41: Intelligent outbound dialing robot
42:自然語言處理模型 42:Natural language processing model
5:第一客戶端裝置 5: First client device
6:第二客戶端裝置 6: Second client device
Claims (5)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW112207050U TWM650999U (en) | 2023-07-07 | 2023-07-07 | Intelligent labeling system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW112207050U TWM650999U (en) | 2023-07-07 | 2023-07-07 | Intelligent labeling system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| TWM650999U true TWM650999U (en) | 2024-02-01 |
Family
ID=90823218
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW112207050U TWM650999U (en) | 2023-07-07 | 2023-07-07 | Intelligent labeling system |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWM650999U (en) |
-
2023
- 2023-07-07 TW TW112207050U patent/TWM650999U/en unknown
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8943004B2 (en) | Tools and methods for determining relationship values | |
| Peyravi et al. | Revolutionised technologies for marketing: Theoretical review with focus on artificial intelligence | |
| US11049137B2 (en) | System and method for human personality diagnostics based on computer perception of observable behavioral manifestations of an individual | |
| US9547832B2 (en) | Identifying individual intentions and determining responses to individual intentions | |
| US20120084120A1 (en) | Survey assessment | |
| US20070259325A1 (en) | Internet-based method and system for human-human question and answer sessions in delayed and real time | |
| CN108521525A (en) | Intelligent robot customer service marketing method and system based on user tag system | |
| US20100241476A1 (en) | Apparatuses, Methods and Systems For A Volunteer Sponsor Charity Nexus | |
| US20180285936A1 (en) | Intelligent visual object management system | |
| WO2007129995A1 (en) | . method and system for providing anonymised consumer opinions to companies | |
| EP2812856A1 (en) | Tools and methods for determining relationship values | |
| KR20070037665A (en) | Instant order system and method using instant messaging, and instant knowledge service system and method using instant messaging | |
| CN111027838A (en) | Crowdsourcing task pushing method, device, equipment and storage medium thereof | |
| US20130262355A1 (en) | Tools and methods for determining semantic relationship indexes | |
| JP7344234B2 (en) | Method and system for automatic call routing without caller intervention using anonymous online user behavior | |
| Shanthi | Customer relationship management | |
| KR20160104462A (en) | Device and System for providing phone number service by providing advertisements using emotion analysis of customer and method thereof | |
| TWM650999U (en) | Intelligent labeling system | |
| US20160189190A1 (en) | Computer implemented system and method for creation of a digital collaborative communication network for generating inquiries and receiving responses over mobile systems to provide customer response data to vendors and distribute selective data over social and other networks | |
| US20160189194A1 (en) | Computer implemented system and method for creation of a digital,collaborative review platform, network and publication | |
| TWI723272B (en) | Real-time customer service system and method thereof | |
| KR101950493B1 (en) | Device and System for providing phone number service by providing customer's emotion to client and method thereof | |
| Brünker et al. | Oiling the ‘Tireless Selling-Machine’–Exploring Requirements for the Deployment of Social Bots in Social Commerce | |
| WO2013119451A2 (en) | Tools and methods for determining relationship values | |
| Ramaul | Role of AI in marketing through CRM integration with specific reference to chatbots |