TWM614548U - Smart chatbot - Google Patents

Abstract

This authoring department provides a smart chat robot, which includes an object database/keyword database, a keyword database, an analysis module, and a search module. The analysis module obtains a plurality of keywords ranked within a predetermined threshold from the keyword database, and analyzes whether the chat conversation content from an external device contains multiple keys ranked within the predetermined threshold based on semantic analysis technology Character. The search module compares the plurality of keywords ranked within the predetermined threshold in the content of the chat conversation and the object characteristics of the object database/keyword database, and finds out the at least one object feature and the content of the chat dialogue The plurality of keywords ranked within the predetermined threshold meet the plurality of objects/keywords within a predetermined condition, and a search result containing a specific number of the plurality of objects/keywords is generated.

Description

Smart chatbot

This creation is about a smart chat robot, especially a search system that finds mutually matching chat objects in predetermined conditions based on a string of chat object search conditions.

There are many considerations for purchasing real estate, such as: price, location, public housing ratio, number of square meters, living functions, object types, etc. Common real estate search methods use the above-mentioned aspects as the criteria for customers to search, for example, the total price of the search is less than one. Tens of millions of objects or searching for objects in two bedrooms and one living room, but the results of the objects displayed by the customer after the aforementioned search are still very messy and disorderly, and it still takes a lot of effort and energy to browse and filter each object one by one. The usual retrieval method cannot fully meet the needs of customers in terms of purchase considerations.

In addition, many customers are not only considering the conditions of the item itself when purchasing real estate, but the priority is whether the item in which the item is located meets their preferences or needs, that is, to find a good house first The concept of finding good objects has gradually become a trend. However, the existing real estate search methods are still based on the conditions of the objects themselves, or they can only query and know which objects are in a certain administrative area. You must select and enter each object category point by point. Knowing the conditions of the item (such as the age of the house, the total price band, etc.), it will take more time for the customer to find the satisfactory item, and the real estate agent cannot smoothly provide the customer with the information of the item sold through the system.

In addition, with the development of semantic recognition technology, more and more More and more companies launch chat bots, hoping to analyze the objects that users want to find through the chat content entered by users. However, users may enter too many keywords, which greatly narrows down the eligible objects, resulting in poor search results.

Because of the position, the applicant, in view of the lack of known technology, invented this creation "intelligent chat robot" to improve the above-mentioned deficiency.

Therefore, there is a need for an intelligent chat robot capable of overcoming the problems of learning skills. In order to achieve the above-mentioned purpose, this authoring department provides a smart chat robot, which includes an object database/keyword database, a keyword database, an analysis module, and a search module. The analysis module obtains a plurality of keywords ranked within a predetermined threshold from the keyword database, and analyzes whether the chat conversation content from an external device contains multiple keys ranked within the predetermined threshold based on semantic analysis technology Character. The search module compares the plurality of keywords ranked within the predetermined threshold in the content of the chat conversation and the object characteristics of the object database/keyword database, and finds out the at least one object feature and the content of the chat dialogue The plurality of keywords ranked within the predetermined threshold meet the plurality of objects/keywords within a predetermined condition, and a search result containing a specific number of the plurality of objects/keywords is generated.

The intelligent chat robot provided by this creation can fully meet the needs of customers in real estate purchase, that is, the intelligent chat robot of this creation searches in an object-first search method, and users input their needs for residential types such as objects. Afterwards, the system presents the result of the collection of objects. Users can first select the objects they are satisfied with, and then browse the objects for sale among them. Since objects can often reflect the living conditions of the objects in them, such as the living conditions of the objects, the priority is given to objects. The search method can effectively allow real estate agents to provide customers with information about the items they sell through the system, thereby increasing the chances of matching between the two parties.

10: Smart chatbot

101: user side

105: Object database/keyword database

106: Plural objects/keywords

110: Analysis Module

120: Search module

130: output module

106A: Object

106B: Object

A01, A02, A03, A04: object features

S01, S02: Saleable items

20: System search page

205: Search type options

206: Search type options

210: search criteria

211: search criteria

215: Search button

220: map block

221: mark

222: mark

230: search result block

231: Project

232: Project

301: Chat conversation content

302a~302c: Keywords

Figure 1 is a schematic diagram showing the intelligent chat robot created by this invention;

Figure 2 is a schematic diagram showing the object database/keyword database of this creation;

Figure 3 is a schematic diagram showing the system search page of this creation;

Figure 4 is a schematic diagram showing the system search page of this creation;

Figure 5 is a schematic diagram showing the system search page of this creation;

Figure 6 is a schematic diagram showing the keyword ranking of this creation;

Figure 7 is a schematic diagram showing the operation of the chatbot created by this creation

This creation will be fully understood by the following examples, so that those who are familiar with the art can complete it, but this creation can also be implemented in other examples. That is, the implementation of this creation is not limited by the following embodiments, but the scope of the patent application for this creation shall prevail.

Please refer to Figure 1, which shows a schematic diagram of the intelligent chatbot created by this invention. In Figure 1, the smart chatbot 10 includes an object database/keyword database 105, an analysis module 110, a search module 120, and an output module 130. The object database/keyword database 105 stores a plurality of Object/keyword 106, and a plurality of objects 106 have at least one object feature (detailed in FIG. 2), and plural keywords 106 (detailed in FIG. 6). The plural keywords in the keyword database 105 are collected from the websites of real estate companies. Every time a user performs a search operation, the search conditions used by the user will be recorded, and each key will be obtained through appropriate weighting processing. Information such as the number of occurrences of the word.

To put it simply, in order to greatly reduce the burden of semantic analysis, but not to oversimplify the results of voice analysis, the creative intelligent chatbot 10 will be based on a continuously accumulated and updated keyword database 10 , To determine which keywords are currently used by most users.

In the smart chatbot 10, the analysis module 110 obtains a plurality of keywords ranked at a predetermined threshold from the keyword database, and analyzes the chat conversation content 301 from the user terminal 101 (such as the seventh (Shown in the figure) whether to include multiple keywords that rank within the predetermined threshold. For example, as shown in Figure 6, when the top three most frequent occurrences of various keywords are administrative districts, budgets, and house types, even if the chat conversation content additionally mentions the number of rooms and the age of the house, it will be affected. Excluded by the system in order to reduce the load on the system while avoiding drastically reducing the number of eligible objects.

As shown in Fig. 7, the search module 120 compares the plurality of keywords 302a~302c ranked within the predetermined threshold in the chat conversation content 301 and the object feature of the object database to find the at least object feature and the The plurality of keywords in the chat conversation content that are ranked at the predetermined threshold meet the plurality of objects within a predetermined condition, and a search result containing a specific number of the plurality of objects is generated.

The interactive chat mode can be conducted in a question-and-answer manner. For example, after the system first asks "the administrative area to be searched", if the user answers "Taipei City", the system can choose to request further explanation of which administrative area. Or when the user answers directly to "Da'an District, Taipei City", the system will not continue to inquire, but will temporarily store the object characteristics about the "administrative area". Of course, the type of temporary storage can be chat, but it is only considered For storage capacity issues, the first chat object condition of the "administrative area" can be converted into a plurality of keywords ranked within the predetermined threshold in the chat dialogue content, and then temporarily stored. Then, complete in order as The condition of the first chat object that is required to be known at least for searching can be reduced, and the problem of too wide search range and too many results can be reduced, and the load of the system can also be reduced.

The above search scenarios include different scenarios such as renting a house, buying a house, selling a house, etc., and in different search scenarios, there will be relative chat recognition phrases for recognition, and the generated chat conversation content ranks within the predetermined threshold. The plural keywords within also roughly belong to the chat recognition phrase, reducing false chat recognition. Furthermore, learning functions can also be incorporated.

For example, in the search context of house purchase and sale, the same first chat object condition, such as Shuanglian, the search context of house purchase is more likely to find objects around Shuanglian MRT station, so the chat conversation content Multiple keywords ranked within the threshold of the reservation are more likely to be Shuanglian MRT station, and in the search context of a house for sale, it is more likely to be looking for objects around Shuanglian Street, so the content of the chat conversation is ranked in the reservation. The multiple keywords within the threshold are more likely to be Shuanglian Street instead of Shuanglian MRT Station.

In order to further expand the search results, the smart chatbot 10 can also automatically expand the search range. After the analysis module 110 receives a plurality of keywords that rank within the predetermined threshold in the chat conversation content sent by the user terminal 101, it sends the received object condition string to the search module 120, and then the search module 120 Compare the object condition string with the object characteristics in the object database/keyword database 105, find the objects that match each other within the predetermined conditions and send the matching objects to the output module 130, and then the output module 130 will search The result is sent back to the user side 101. Preferably, the search module 120 finds objects that match each other within a predetermined condition and generates a search result containing a specific number of objects, where the specific number can be zero or an integer greater than zero.

Please refer to Figure 2, which shows a schematic diagram of the object database/keyword database of this creation. In Figure 2, the object database/keyword The database 105 stores the object 106A, for example, a suitable house, and the object 106B, for example, a building. For example, the objects can be military villages, built-up houses, national houses, collective residential areas, suitable houses, social houses, or apartments within a specific range, mansions, residential buildings, suites, touring houses, or other dwellings. Type. Object 106A has object feature A01, for example, the age of the house is 15 years, object feature A02, for example, the number of households in the community to which the object belongs, 20 households, object feature A03, for example, the living function is a store, and object feature A04 is, for example, a public swimming pool, and object 106A is also It has a saleable item S01 and a saleable item S02. The object 106B has an object feature A01, for example, a house age of 15 years, an object feature A02, for example, the number of households in the community to which the object belongs, 15 households, and an object feature A03, for example, a living function as a market, where the saleable items of the object 106B are zero. For example, the feature of an object may be the number of houses, the age of the house, the public status, the number of households in the community to which the object belongs, the living function, the number of floor space, the orientation of the house, and the number of floors of the house. Preferably, the object database/keyword database 105 can store data of one or more objects. Preferably, the object may have one or more object characteristics. Preferably, the saleable items of the item may be zero or an integer greater than zero.

The correspondence between the object 106A and the saleable item S01 and the saleable item S02 needs to be completed by some means, because there is no public information to find out the correspondence relationship. Specifically, based on the township, urban area, location of the land section, or building address, total land transfer area square meters, urban land use zoning, transaction year and month, number of transaction buildings, transfer level, total number of floors , Building type, main purpose, main building materials, building completion year and month, total area of the building transferred square meters, current state of the building, whether there is a management organization, the total price, the unit price per square meter or the name of the object, etc., you can choose All, or select several from them, and construct the object characteristics of a plurality of objects, and then disassemble the transaction data into a plurality of fields based on this, and provide search and comparison purposes. Then, according to the housing conditions of the saleable item S01 and the saleable item S02, it means the location in the house condition The address, total number of floors, and construction completion year and month are compared with the object characteristics of most objects in the object database/keyword database to search for at least the corresponding housing conditions that meet the saleable object S01 and the saleable object S0 A specific object information, and the establishment of the relationship between the object 106A and the saleable item S01 and the saleable item S02 is completed.

Please continue to refer to Figures 1 and 2, the search module 120 compares the received object condition string with the object characteristics of a plurality of objects/keywords 106 in the object database/keyword database 105, and finds out where Objects that meet each other within the predetermined conditions, for example, if the object condition string is 15 years old, the search module 120 compares the object characteristics in the object database/keyword database 105 and finds the object characteristics of the object 106A and the object 106B A01 is the house age of 15 years, so the objects that meet the matching conditions are the object 106A and the object 106B, and then the search module 120 sends the object 106A and the object 106B to the output module 130 to generate search results.

In this creation, the search module 120 will perform comparisons within predetermined conditions, and the predetermined conditions include (1) the object condition string is a numeric value, and the housing group characteristics are the same as the object condition string, and the search results show objects that completely match; (2) The object condition string is a numeric value. If the housing group characteristics are different from the object condition string, the difference range is the range of the object condition string plus or minus the predetermined value. The search result shows some objects that match; (3) The object condition string is If the text, housing group characteristics and the object condition string are the same, the search results show objects that completely match; and (4) The object condition string is text, and the housing group characteristics are different from the object condition string, then the search results show the housing group characteristics and objects Objects with similar condition strings. For example, if the object condition string is 20 households in the community to which the object belongs, the search module 120 will compare the object characteristics in the object database/keyword database 105 with the value 20 plus or minus 5 in the search range of 15-25 , And then find out the object 106A with 20 households in the community where the object belongs and 106B with 15 households in the community where the object belongs. If the object condition string is life The function is the market, and the search module 120 not only finds the object 106B with the housing group characteristics having the market, but also uses the similar characteristics to find the object 106A with the housing group characteristics having the store. Among them, markets, stores, etc. belong to similar conditions and need to be summarized in advance and build an object database/keyword database.

Please refer to Figure 3, which shows a schematic diagram of the system search page of this creation. In FIG. 3, the system search page 20 includes search type options 205-206, search conditions 210-211, and a search button 215. The user selects the type option and enters the search condition and clicks the search button 215 to send the object condition string to the analysis module of the authoring system. Preferably, the search type option 205 and search condition 210 are required input data, and the search type option 206 and search condition 211 are unnecessary input data. If the user wants to input multiple search data, he can click on the system search page 20 "+" to generate new search type options and search criteria fields.

Please refer to Figure 4, which shows a schematic diagram of the system search page of this creation. In Figure 4, the system search page 20 further includes a map block 220 and a search result block 230. The map block 220 displays the location of the search result object on the map, for example, the object 106A is indicated by the mark 221 and the object 106B It is represented by the mark 222, and the search result block 230 displays the name of the object of the search result, for example, the object 106A is represented by the item 231 and the object 106B is represented by the item 232. For example, after the user sends out the object condition string such as the age of the house from 15 years to the analysis module of the creative system, the object database/keyword database will be compared with the object database/keyword database through the search module and the output module will have Housing group feature A01 is, for example, a 15-year-old object 106A, such as a suitable house, and object 106B, such as a mansion, sent back to the system search page 20, where the search module will confirm the matching object after finding the matching object Whether the object has a saleable object, if the matching object does not have a saleable object, it will be listed as a non-conforming object and removed from the comparison result, that is, the search result does not show a match Objects with comparison conditions but without any saleable item data. In this example, object 106A has saleable item S01 and saleable item S02, and object 106B has saleable item S01, so they all have saleable items and meet the comparison conditions In addition, the search module will also sort the objects in the search results by at least the number of saleable items in the objects that meet the matching conditions. For example, item 106A has more saleable items than item 106B, so The order of the object 106A is higher than that of the object 106B, so that the user can know which objects meet the search criteria and have saleable objects, and can preferentially select objects with more saleable objects.

Please refer to Figure 5, which shows a schematic diagram of the system search page of this creation. Fig. 5 is a continuation of Fig. 4. After the user obtains the preliminary search results, the user performs operations related to one of the second chat object conditions of the at least one object feature, and the analysis module 110 uses the relative to the search according to the search context. The chat identification phrase of the context, the second chat object condition is converted into a second object condition string, and the second object condition string is used to search again in the preliminary search results, where the data of the preliminary search results has been stored in On the client side, the re-searching process will only read the client’s cache, and will not re-access the object database/keyword database to reduce the burden on the object database/keyword database.

In more detail, the search module compares the second object condition string with at least one object feature of the object that meets the matching conditions in the preliminary search results, and finds out that at least one housing group feature and the second object condition string are predetermined Objects that match each other in the conditions, and generate a search result that includes one of the objects that meet the matching conditions. If the search module compares the second object condition string with at least one object feature of the object that meets the matching conditions in the search result Find out the objects that at least one housing group feature and the second object condition string match each other within the predetermined conditions, then the search module removes the non-conforming objects from the search results and presents the matching objects in the re-search result . For example, the user sends a second object condition string such as swimming pool, and In the preliminary search result, only the object 106A has housing characteristics A04, such as a public swimming pool, so the system search page 20 removes the object 106B from the map block 220 and the search result block 230 according to the second object condition string, leaving only The matching object 106A is the re-search result. Preferably, if the search module compares the second object condition string with the object feature matching the comparison condition in the preliminary search result, it does not find that at least one object feature and the second object condition string match each other within the predetermined condition The search result is the same as the preliminary search result.

This creation is disclosed with the above-mentioned preferred embodiments and examples as references, and readers should understand that these examples are used for description rather than limitation. Anyone who knows this skill can make various combinations and modifications without departing from the spirit and scope of the present invention, and it should still fall within the scope of the patent of the present invention.

10: Smart chatbot

101: user side

105: Object database/keyword database

106: Plural objects/keywords

110: Analysis Module

120: Search module

130: output module

Claims (9)

A smart chat robot, including: An object database/keyword database configured to store a plurality of objects/keywords, wherein each of the objects includes at least one object feature; A keyword database configured to store a plurality of keywords, wherein each keyword includes at least one occurrence; An analysis module obtains a plurality of keywords ranked within a predetermined threshold from the keyword database, and analyzes whether the content of a chat conversation from a user terminal contains rankings within the predetermined threshold based on semantic analysis technology Plural keywords within; and A search module compares a plurality of keywords ranked within the predetermined threshold in the content of the chat dialogue and the object characteristics of the object database/keyword database, and finds out the at least one object characteristic and the chat dialogue The plurality of keywords ranked within the predetermined threshold in the content match the plurality of objects/keywords within a predetermined condition, and a search result containing a specific number of the plurality of objects/keywords is generated. The smart chat robot described in item 1 of the scope of patent application, wherein the mutual matching within the predetermined conditions means that the plurality of keywords in the content of the chat conversation that are ranked within the predetermined threshold fall in the at least one residence Within a range of differences in group characteristics. For the smart chat robot described in item 2 of the scope of patent application, the mutual compliance within the predetermined conditions includes: (1) The plurality of keywords in the content of the chat conversation that rank within the predetermined threshold are numerical values, The at least one housing group feature is the same as a plurality of keywords ranked within the predetermined threshold in the chat dialogue content, and the search results show objects that completely match; (2) the chat dialogue content ranks within the predetermined threshold The plural keywords of are numeric values, which to If the feature of one less housing group is different from the plural keywords in the chat conversation content that rank within the predetermined threshold, the range of the difference is the plural keywords in the chat conversation content that rank within the predetermined threshold plus or minus one reservation The range of values, the search result shows some objects that match; (3) The plurality of keywords in the chat conversation content that rank within the predetermined threshold are text, and the at least one housing group feature and the chat conversation content rank in The plurality of keywords within the predetermined threshold are the same, and the search results show objects that completely match; and (4) the plurality of keywords in the chat conversation content that rank within the predetermined threshold are text, and the at least one housing group The feature is different from the plurality of keywords in the chat conversation content that rank within the predetermined threshold, then the search result shows the at least one housing group feature and the plurality of keywords in the chat conversation content that rank within the predetermined threshold Objects that are similar in part. The smart chat robot described in item 3 of the scope of patent application, wherein the search module is configured to compare a second object condition string with the plurality of objects/keywords in the search result that meet the matching conditions For the at least one object feature, find the plurality of objects/keywords that match the at least one housing group feature and the second object condition string within the predetermined condition, and generate the plurality of objects/keywords that meet the matching condition One of the objects/keywords then search the results, where if the search module compares the second object condition string with the at least one object feature of the plurality of objects/keywords in the search result that meets the matching conditions, it finds If the at least one housing group feature and the second object condition string match each other within the predetermined condition, the search module removes the object that does not match from the search result and presents the match in the re-search result的物件。 The object. The smart chat robot described in item 4 of the scope of patent application, wherein the condition string of the second object is read from the cache data, and the content of the chat conversation is ranked within the predetermined threshold Data generated by multiple keywords of without re-accessing the object data Database/Keyword database data. For example, in the smart chat robot described in item 4 of the scope of patent application, the object database/keyword database is further configured to store a saleable object data of each object and the search module is configured to match The plurality of objects/keywords of the comparison condition and the saleable object data that they have, and the number of the plurality of objects/keywords that meet the comparison condition of the saleable object data is sorted by the most at least The plurality of objects/keywords in the search result and the re-search result. For example, in the smart chat robot described in item 6 of the scope of patent application, if the plurality of objects/keywords that meet the matching conditions do not have any saleable object data, the search result and the re-search result do not appear to match The plural objects/keywords of the matching condition. For example, the smart chat robot described in item 3 of the scope of patent application, wherein the feature of the at least one object is selected from the group consisting of: the number of houses, the age of the house, the status of a public establishment, the number of households in the community to which an object belongs, a living function, and a house A group consisting of the number of pings, the orientation of a house, the number of floors of a house, and their combination. For example, the smart chat robot described in item 1 of the scope of patent application, wherein the plural objects/keywords are military villages, rebuilt houses, national houses, collective settlements, suitable houses, social houses, or within a specific range Of apartments, mansions, residential buildings, suites, and open houses.

Images