Auxiliary support system for project management systems

Abstract

The present invention relates to an auxiliary support system for project management systems, wherein the project management system has a predefined format. The auxiliary support system allows the input of sentence groups via voice or text and enables the training of the sentence groups to generate project management commands. The auxiliary support system includes a keyword extraction unit that extracts keywords. Subsequently, a text analysis unit analyzes the extracted keywords based on various preset attribute categories. A comparison unit utilizes an artificial intelligence language model to compare the keywords classified by the text analysis unit into respective attributes with multiple formats, thereby generating project management commands. A format mapping unit can generate code corresponding to the format of the project management system. Through this code, tasks can be added or schedules modified within the project management system. This achieves integration with multiple applications, thereby enhancing work efficiency.

Description

[Technical Field]

[0001] The present invention relates to an auxiliary support system for a project management system, particularly an auxiliary support system that can connect to multiple applications and can be directly input via voice or text. [Previous Technology]

[0002] More and more people are now engaged in project-based work, such as self-media practitioners, cultural and creative market vendors, online advertising practitioners, software engineers, developers, turnkey construction companies, or highly focused jobs such as designers, advertising planners, and creative directors. These groups often need a lot of time to conceive and create, but at the same time they are troubled by tedious project details, such as special sales events on January 1st, contacting event venues, contacting catering suppliers on the day, and other errands unrelated to creativity. This will cause project implementers to be distracted and unable to focus on the core creative idea, and will also sharply reduce their job satisfaction and productivity.

[0003] Currently, the most commonly used project management tools (such as Asana, Trello, Monday.com) require project executors to go to their respective pages to input text. However, in addition to the inconvenience of switching between multiple project management tools, it is also extremely cumbersome to input content into separate fields within a single project management tool. In other words, in addition to producing content related to creative ideas, these project executors also need to use one or more different types of project management tools to record to-do items. The way of recording requires a period of focused time to fill in each field, which is very energy-consuming for project executors.

[0004] Artificial intelligence products or programs are becoming increasingly advanced. For example, some people use CHAT GPT as an assistant to record their personal schedules, or ask it to extract summaries from messy articles. However, these ways of making life easier can only exist within the application. In other words, current project implementers need to switch between various applications to meet different needs, which is very energy-consuming.

[0005] Therefore, in order to enable project implementers to use project management tools more conveniently, intuitively, and quickly, and at the same time enable project implementers to focus on a single project management tool to improve productivity, it is necessary to develop ideal technical methods to solve the above problems. [Summary of the Invention]

[0006] The purpose of this invention is to provide an auxiliary support system for a project management system, which combines different project management tools to simplify and integrate the task management process into one interface, and allows input via either voice or text, making it easier for users to operate.

[0007] This invention relates to an auxiliary support system for a project management system. The project management system has a plurality of predetermined formats. The format content can be fields related to filling in tasks, such as Title, Content, Date, and Project. The auxiliary support system can generate project management instructions by training sentence groups. The auxiliary support system includes a keyword extraction unit, a text analysis unit, a comparison unit, and a format correspondence unit.

[0008] The keyword extraction unit has a natural language instruction set, and the keyword extraction unit extracts multiple keywords through the text sentence group of the natural language instruction set.

[0009] In the text analysis unit, the keywords will correspond to a variety of preset attribute categories. The text analysis unit classifies the keywords extracted by the keyword extraction unit according to different attribute categories. In other words, after the keyword extraction unit extracts the keywords, the text analysis unit will analyze the various preset attribute categories corresponding to the keywords.

[0010] The comparison unit uses an artificial intelligence language model to compare the keywords of each attribute classified by the text analysis unit with multiple formats to generate project management instructions.

[0011] The format matching unit can generate program code that corresponds to the format of the project management system by using an artificial intelligence language model for project management instructions. The program code can be used to add tasks or modify schedules in the project management system.

[0012] The natural language instruction set in the assisted support system is formed by semantically and classifying at least two types of corpus data. Since the corpus data can come from different sources, such as operation records of project management tools (e.g., Asana), public API documentation, existing instruction sets, or user behavior analysis, different instructions can be matched with corresponding project operations through semantic and classification annotation, enabling the assisted support system to learn how to convert natural language into specific operations. The specific difference between semantic and classification annotation is that semantic annotation is used to annotate the semantically corresponding operation behavior of natural language, while classification annotation classifies different instructions into specific task types.

[0013] The artificial intelligence language model in the auxiliary support system adopts the T5 (Text-To-Text Transfer Transformer) model, which performs well in generative tasks.

[0014] The accessibility support system is coupled to the project management system such as Asana and the speech-to-text device such as Google Cloud Speech or Azure Speech via API (application programming interface). The sentence is formed by either text or speech. The speech-to-text device is used to convert speech into sentence. In other words, users can input either speech or text into the accessibility support system. If it is speech, it can be converted into text through Google Cloud Speech or Azure Speech for subsequent use.

[0015] The auxiliary support system further includes an application unit, a storage unit and an output unit. When the user inputs text, the application unit stores the text in the storage unit. The application unit obtains the format for the comparison unit to compare the keywords with the format to generate project management instructions. The format matching unit outputs the project management instructions to the format corresponding to the project management system, and then the output unit reports back to the user.

[0016] When a user inputs with voice in the auxiliary support system, the voice is transmitted to the speech-to-text device through the application unit. After the speech-to-text device generates text, it stores the text in the storage unit. The application unit obtains the format for the comparison unit to compare the keywords with the format in order to generate project management instructions. The format matching unit outputs the project management instructions to the corresponding format of the project management system, and then the output unit reports back to the user.

[0017] Therefore, by utilizing the auxiliary support system for project management systems provided by the present invention, users can input via voice or text. If it is voice input, the auxiliary support system first converts the voice into text using a speech-to-text device (if it is text input, it does not need to use a speech-to-text device). Then, the keyword extraction unit extracts keywords, and the project management instructions are output to the corresponding format of the project management system through the text analysis unit, comparison unit, and format matching unit. Finally, the output unit reports back to the user. In this way, multiple digital tools can be coupled through API in the same system. As the demand for distributed work and remote collaboration increases, it can provide cross-platform and multi-device synchronization functions to improve work efficiency.

[0018] The advantages and spirit of the present invention can be further understood from the following detailed description of the invention and the accompanying drawings.

Implementation Method

[0019] The purpose of this invention is to provide an auxiliary support system for a project management system, which combines different project management tools to simplify and integrate the task management process into one interface, and allows input via either voice or text, making it easier for users to operate.

[0020] Please refer to Figure 1 and Figure 2. Figure 1 is a functional component diagram of the auxiliary support system of the present invention, and Figure 2 is a schematic diagram of the actual operation of the auxiliary support system of the present invention. Figure 2 is also the first embodiment. First, the present invention is an auxiliary support system 100 for a project management system 300. The project management system 300 has a predetermined plurality of formats 302. The content of the format 302 can be fields related to filling in tasks, such as Title, Content, Date, Project. The speech-to-text device 400 is used to convert speech into text. The auxiliary support system 100 can train the sentence group 10 to generate project management instructions 62. In the first embodiment, the auxiliary support system 100 is coupled to the project management system 300, such as Asana, and the speech-to-text device 400, such as Azure Speech, via APIs. The sentence group 10 is formed by either text or speech. That is, the user 1 can input to the auxiliary support system 100 in the form of speech or text. If it is speech, it can be converted into text through Azure Speech for subsequent use.

[0021] The auxiliary support system 100 includes an application unit 20, a storage unit 30, a keyword extraction unit 40, a text analysis unit 50, a comparison unit 60, a format matching unit 70, and an output unit 80.

[0022] The application unit 20 will first obtain the format 302 of the project management system 300 for subsequent comparison between the auxiliary support system 100 and the project management system 300, and will generate project management instructions 62.

[0023] The storage unit 30 is where the user 1 inputs text sentences 10. If the input is text, the text can be stored in the storage unit 30 through the application unit 20. Alternatively, if the input is voice, the voice is converted into text through the voice-to-text device 400 and the text is stored in the storage unit 30.

[0024] The keyword extraction unit 40 has a natural language instruction set 42, and the keyword extraction unit 40 extracts multiple keywords 16 through the natural language instruction set 42 and the self-text sentence group 10.

[0025] In the text analysis unit 50, the keyword 16 will correspond to a variety of preset attribute categories. The text analysis unit 50 classifies the keyword 16 extracted by the keyword extraction unit 40 according to different attribute categories. In other words, after the keyword extraction unit 40 extracts the keyword 16, the text analysis unit 50 will analyze the various preset attribute categories corresponding to the keyword 16.

[0026] The comparison unit 60 compares the keywords 16 of each attribute classified by the text analysis unit 50 with a plurality of formats 302 through the artificial intelligence language model 200 to generate project management instructions 62.

[0027] The format corresponding unit 70 can generate program code 72 that corresponds to the format 302 of the project management system 300 through the artificial intelligence language model 200 for the project management instruction 62. The program code 72 can be used to add tasks or modify schedules in the project management system 300.

[0028] After the format corresponding unit 70 outputs the project management instruction 62 to the format 302 corresponding to the project management system 300, the output unit 80 reports it back to the user 1.

[0029] The natural language instruction set 42 in the keyword extraction unit 40 of the auxiliary support system 100 is formed by semantic annotation and classification annotation of at least two kinds of corpus data 44. Since the corpus data 44 can come from different sources, such as operation records of project management tools (e.g., Asana), public API documents, existing instruction sets, or user behavior analysis, after semantic annotation and classification annotation, different instructions can be matched with corresponding project operations, so that the auxiliary support system 100 can learn how to convert natural language into specific operations. The specific difference between semantic annotation and classification annotation is that semantic annotation is used to annotate the semantic corresponding operation behavior of natural language, while classification annotation is to classify different instructions into specific task types.

[0030] The artificial intelligence language model 200 in the auxiliary support system 100 adopts the T5 (Text-To-Text Transfer Transformer) model. The T5 model performs well in generative tasks. Alternatively, the BERT (Bidirectional Encoder Representations from Transformers) model, which is more suitable for comprehension tasks, can also be adopted. By training the artificial intelligence language model 200 multiple times, the auxiliary support system 100 can more easily understand the meaning of the user's sentence group 10.

[0031] When user 1 inputs by voice, the voice can be transmitted to the speech-to-text device 400 through the application unit 20. After the speech-to-text device 400 generates text, it stores the text in the storage unit 30. The application unit 20 obtains the format 302 for the comparison unit 60 to compare the keyword 16 with the format 302 to generate the project management instruction 62. The format matching unit 70 outputs the corresponding project management instruction 62 to the project management system 300 through the program code 72 and then reports it to user 1 through the output unit 80. At this time, user 1 can see the content filled in the format 302.

[0032] If user 1 inputs text, the speech does not need to be transmitted to the speech-to-text device 400 through application unit 20. That is, the text can be directly stored in storage unit 30. After application unit 20 obtains format 302 and comparison unit 60 compares keyword 16 with format 302, project management instruction 62 will be generated as mentioned above. Format corresponding unit 70 outputs the corresponding project management instruction 62 to the project management system 300 through process code 72. Then, output unit 80 reports the result to user 1. At this time, user 1 can see the content filled in format 302.

[0033] Please refer to Figure 3, which is a document diagram of the project management instruction parameters in the first embodiment. The figure shows the specific document description of the project management instruction 62. First, it is necessary to generate JSON parameter output. For example, specific task instructions, such as "add task" or "modify schedule", are extracted from the Chinese sentence group 10 output by the auxiliary support system 100. At this time, it is necessary to format the JSON parameters. In other words, the instructions in the sentence group 10 output by the auxiliary support system 100 are converted into a standardized JSON format so as to interact with the project management system 300, such as Asana. These data may include: specified project tasks, task descriptions, schedules, etc.

[0034] The comparison unit 60 first uses the artificial intelligence language model 200, such as the T5 model, to compare the keywords 16 of each attribute classified by the text analysis unit 50 with the multiple formats 302 in the project management system 300. In other words, the project management instruction 62 is generated by first formatting the JSON parameters and then converting the sentence group 10 instruction into a standardized JSON format.

[0035] Please refer to Figure 4, which is a schematic diagram of the code documentation in the first embodiment. The figure shows an API document conforming to the JSON format. After the subsequent project management instruction 62 is obtained, an HTTP POST request can be made using the pre-designed project management instruction 62, i.e., JSON format data, according to the API document of the project management system 300. This is the HTTP method for sending data to the server. In this way, the results of the auxiliary support system 100 can be sent to the project management system 300 for storage through API coupling.

[0036] Please refer to Figure 5, which is a flowchart of the operation method of the auxiliary support system 100 of the present invention. As mentioned above, the present invention is an auxiliary support system 100 for a project management system 300, and the project management system 300 has a predetermined plurality of formats 302, wherein the auxiliary support system 100 can be API-coupled to the project management system 300, such as Asana, and the speech-to-text device 400, such as Azure Speech.

[0037] Step 1 S01: User 1 inputs sentence group 10 into application unit 20 of auxiliary support system 100, wherein sentence group 10 can be input by voice or text.

[0038] Step 2 S02: If the input sentence group 10 is voice input, the voice is first converted into text by API coupling to the speech-to-text device 400 and then stored in the storage unit 30; if the input sentence group 10 is text input, it is directly stored in the storage unit 30.

[0039] Step 3 S03: Keyword extraction unit 40 extracts multiple keywords 16 through natural language instruction set 42 self-text sentence group 10.

[0040] Step 4 S04: Text analysis unit 50 classifies the keywords 16 extracted by keyword extraction unit 40 according to different attribute categories.

[0041] Step 5 S05: The comparison unit 60 compares the keywords 16 of each attribute classified by the text analysis unit 50 with a plurality of formats 302 through the artificial intelligence language model 200 to generate the project management instruction 62.

[0042] Step Six S06: The format corresponding unit 70 then uses the artificial intelligence language model 200 to generate program code 72 that corresponds to the format 302 of the project management system 300 from the project management instruction 62. Then, through API coupling, the result from the auxiliary support system 100 is sent to the project management system 300 for storage. In other words, the final text will be stored in format 302.

[0043] Step 7 S07: Finally, the output unit 80 reports back to the user 1.

[0044] Therefore, by utilizing the auxiliary support system 100 for project management system 300 provided by the present invention, user 1 can input via voice or text. If it is voice input, the auxiliary support system 100 first converts the voice into text through the speech-to-text device 400 (if it is text input, the speech-to-text device 400 is not required). Then, the keyword extraction unit 40 extracts the keyword 16. The project management instruction 62 is output to the project management system 300 in the corresponding format 302 through the text analysis unit 50, the comparison unit 60, and the format matching unit 70. Finally, the output unit 80 reports back to user 1. In this way, multiple digital tools can be coupled through API in the same system. As the demand for distributed work and remote collaboration increases, it can provide cross-platform and multi-device synchronization functions to improve work efficiency.

[0045] The detailed description of the preferred embodiments above is intended to more clearly describe the features and spirit of the present invention, and is not intended to limit the scope of the present invention with the preferred embodiments disclosed above. On the contrary, the aim is to cover various modifications and equivalent arrangements within the scope of the patent claims to be made by the present invention. [Simplified Explanation of the Diagram]

[0046] Figure 1 is a functional element diagram of the auxiliary support system of the present invention; Figure 2 is a schematic diagram of the actual operation of the auxiliary support system of the present invention; Figure 3 is a document schematic diagram of the project management instruction parameters in the first embodiment; Figure 4 is a document schematic diagram of the program code in the first embodiment; and Figure 5 is a flowchart of the method for operating the auxiliary support system of the present invention.

Claims

1. An auxiliary support system for a project management system, the project management system having a predetermined plurality of formats, the auxiliary support system enabling a sentence group to generate a project management instruction, the auxiliary support system comprising: a keyword extraction unit having a natural language instruction set, the keyword extraction unit extracting multiple keywords from the sentence group through the natural language instruction set; a text analysis unit, the keywords corresponding to multiple preset attribute categories, the text analysis unit classifying the keywords extracted by the keyword extraction unit according to different attribute categories; a comparison unit, the comparison unit comparing the keywords classified by the text analysis unit for each attribute with the plurality of formats through an artificial intelligence language model to generate the project management instruction; and a format matching unit, generating program code corresponding to the format of the project management system for the project management instruction through the artificial intelligence language model, wherein the artificial intelligence language model adopts the T5 (Text-To-Text Transfer Transformer) model.

2. The auxiliary support system as described in claim 1, wherein the natural language instruction set is formed by semantically and classifying at least two corpus data.

3. The auxiliary support system as described in claim 1, wherein the auxiliary support system is coupled to the project management system and a speech-to-text device via an API (application programming interface), and the sentence group is formed by either text or speech, wherein the speech-to-text device is used to convert the speech into the sentence group.

4. The auxiliary support system as described in claim 3, wherein the auxiliary support system further includes an application unit, a storage unit and an output unit. When a user inputs the text, the application unit stores the text in the storage unit. The application unit obtains the format for the comparison unit to compare the keyword with the format to generate the project management instruction. The format matching unit outputs the project management instruction to the format corresponding to the project management system, and the output unit reports it back to the user.

5. In the auxiliary support system described in claim 3, when a user inputs voice, the voice is transmitted to the speech-to-text device through the application unit. After the speech-to-text device generates text, it stores the text in the storage unit. The application unit obtains the format for the comparison unit to compare the keyword with the format to generate the project management instruction. The format matching unit outputs the project management instruction to the format corresponding to the project management system, and the output unit reports it back to the user.

Images