Features
Features:

Product Tour >

Edraw AI >

Paid Plans:

Individuals >

Business >

Eduaction >
Resources
Blog

History

How-tos & Tips

Discovery

Biography

Business Analysis

Examples

AI concept Map

Free AI Mind Map Generator

Onenote Mind Map

Bcg Matrix Examples

Nike Marketing Strategy

Unilever SWOT Analysis

Make Mind Maps in Google Docs

Guide

FAQs

What's New

Resource Center
Templates
All Templates

Brain Storming Templates

Strategy and Planning Templates

Project Management Templates

Product Management Templates

Human Resources Templates

Agile Workflow Templates

Marketing Templates

Education Templates

Fun and Games Templates

User Gallery
Download
Pricing
Enterprise

MindMap Gallery Use DeepSeek to build a personal knowledge base

Use DeepSeek to build a personal knowledge base

This is a mind map about using DeepSeek to build a personal knowledge base. The main contents include: instructions, knowledge base use, knowledge base construction, and preparation work.

Edited at 2025-02-11 11:27:52

Gavinnblake

Recent works View more works>>

Use DeepSeek to build a personal knowledge base

Gavinnblake

Recent works View more works>>

Recommended to you
Outline

The Eight Laws of Making Money
- 191
- 5
- 1
Fiona_
How to Make Money in 21 Century
- 123
- 6
Fiona_
The logic of making money
- 27
- 1
MM
Why use DeepSeek to build a personal knowledge base
- 49
- 1
Gavinnblake
Guide to building a personal knowledge base with DeepSeek
- 2.1k
- 41
Gavinnblake
How to make more money with DeepSeek
- 21
- 2
- 1
Mason·Carter
30 High Potential Money-making Tracks Based on DeepSeek (Artificial Intelligence Technology) (Chinese)
- 22
- 1
- 1
Eeeelizabeth

Use DeepSeek to build a personal knowledge base

Preparation

Content preparation

Prepare your own knowledge base content, such as manuals, etc.

Preparation

Content preparation

Prepare your own knowledge base content, such as manuals, etc.

Tool preparation

Download and install the ollama tool

Use ollama to download and install deepseek model, select 1.5b or 7b parameters

Download the embedded model bge-m3

Download and install the AI visualization tool cherryStudio

Create a knowledge base

Generate a Q&A

You can use bean buns to analyze the instructions and generate a Q&A

Upload knowledge base

Copy and paste the generated Q&A pair onto the TXT text, and then drag the TXT text into the Secret Tower AI

Verification Knowledge Base

After waiting for the vectorization to be completed, you can use the Seita AI to perform Q&A on the uploaded knowledge base.

Use knowledge base

Select a workspace

Choose your own workspace

Q&A mode

When asking, you can choose deep thinking or concise answering methods

Tool preparation

Download and install the ollama tool

step

Download and install the ollama tool

Visit the official website of ollama to download and install

Download and install the DeepSeek model

Use ollama to download and install the DeepSeek model, you can choose 1.5b or 7b parameters

Download the embedded model bge-m3

Download embedded model bge-m3 using ollama

Download and install the AI visualization tool cherryStudio

Visit the official website of cherry-ai to download and install

Upload knowledge base

Upload the knowledge base file to cherryStudio

Set up ollama

Set ollama in cherryStudio interface, load models DeepSeek and bge-m3

Enter the dialogue interface

In the cherryStudio interface, enter the dialogue interface, select the DeepSeek model and knowledge base, and ask questions

Things to note

DeepSeek is an inference model and is good at processing in the field of quantitative trading. It may not work well for other knowledge processing.

Use ollama to download and install deepseek model, select 1.5b or 7b parameters

step

Download and install Ollama

Ollama supports a variety of large models and is a tool needed to deploy DeepSeek.

Download the DeepSeek model

Enter Ollama official website, find Models, and select the corresponding DeepSeek model to download

Chatbox front-end construction

Through Chatbox, web pages or client operations can be implemented, making it convenient to use DeepSeek

Things to note

The larger the model, the higher the configuration of computer memory, graphics card, etc.

After deploying DeepSeek, some data that is inconvenient to be disclosed, such as experimental data, can be processed by deploying large models to the local level, so there is no need to worry about data leakage.

Download the embedded model bge-m3

Preparation tools

Download and install Ollama

Visit the official website of ollama to download and install

Download and install the DeepSeek model

Search for the DeepSeek model on the official website of ollama, select the appropriate version according to the computer configuration to download and install it

Download and install the AI visualization tool CherryStudio

Visit CherryStudio official website to download and install

Download the bge-m3 embedding model

Download the bge-m3 embedding model using the ollama tool

Download and install the AI visualization tool cherryStudio

Preparation

Download and install tools

Download and install the ollama tool

Download and install the CherryStudio client

Configuring API Keys in CherryStudio

Knowledge Base Construction

Add knowledge base files

Add supported file formats, such as pdf, docx, pptx, xlsx, txt, md, mdx, etc.

Add the entire folder directory, and files in the supported format will be processed automatically

Add data from multiple sources such as URL links, site maps, plain text notes, etc.

Wait for vectorization to complete

When the display vectorization has been completed, the processing of the file is completed

Use knowledge base

Search the Knowledge Base

Enter the query content to search the knowledge base and obtain relevant results and matching scores

Improve search accuracy

Accurate keyword selection

Multi-keyword combination

Use boolean operators

Limited search range

Clear problem statement

Utilize categories and tags

Troubleshooting items

Verification and adjustment

Knowledge Base Construction

Content processing

Through bean buns, analyze the content of the knowledge base and generate Q&A.

Use DeepSeek to build a personal knowledge base

Preparation

Download and install necessary tools such as ollama

Create a knowledge base

Deploy a local knowledge base with DeepSeek

Upload the required information, such as the instruction manual, as the content of the knowledge base

Generate a Q&A

Generate Q&A with DeepSeek

Using DeepSeek's intelligent system to understand, analyze and generate information

Enter the question and DeepSeek will generate the corresponding answer

Copy and paste the generated Q&A pair onto the TXT text

Prepare the knowledge base content

Prepare your own knowledge base content, which can be in the form of instructions, etc.

Content processing

Analyze the knowledge base content through tools such as beanbao to generate Q&A and Q&A

Question and answer formats can help DeepSeek better learn knowledge bases and generate vector databases

Export Q&A

Copy and paste the generated Q&A pair onto the TXT text

Vectorization processing

Drag TXT text into tools such as Seita AI and other tools, and vectorized by the DeepSeek model

The time of vectorization depends on the content size and content complexity, as well as the parsing speed of the model

Use knowledge base

After vectorization is completed, you can use the Seita AI to perform Q&A on the uploaded knowledge base.

Knowledge Base Vectorization

Drag the TXT text into the Secret Tower AI and let the deepseek model vectorize the knowledge base

Go to the official website of DeepSeeKV3, register an account and obtain APIKey

Select Knowledge Base Management Tool

You can use free tools such as AnythingLLM to manage your knowledge base

Configure APIKey

Fill in the obtained DeepSeeKV3 APIKey into the corresponding settings

Add knowledge base files

Click the button to add file, select the supported file format and open it. The system will automatically perform word segmentation and index creation.

You can also add the entire folder directory, and files that support formats under this directory will be processed automatically.

Or add data from multiple sources such as URL links, site maps, plain text notes, etc.

Wait for vectorization to complete

When the display vectorization has been completed, the processing of the file is completed

Get started with the knowledge base

By entering the query content, you can search the knowledge base and obtain relevant results and matching scores

Improve search accuracy

Accurate keyword selection

Multi-keyword combination

Use boolean operators

Limited search range

Clear problem statement

Utilize categories and tags

Troubleshooting items

Verification and adjustment

Knowledge Base Use

Q&A function

Select your own workspace in Seta AI and ask questions

Build a personal knowledge base

step

Download and install the ollama tool: https://ollama.com

Use ollama to download and install the deepseek model: select 1.5b or 7b, or you can run deepseek directly under ollama, ollamarundeepseek-r1:1.5b

Download the embedded model bge-m3: ollamapullbge-m3

Download and install the AI visualization tool cherryStudio: https://cherry-ai.com/download

Upload knowledge base information: for example, "Mes Manual"

Set ollama in cherryStudio interface: load model deepseek and bge-m3

Enter the dialogue interface: select deepseek model, select mes knowledge base, and ask questions

Ask a question

Question method

Through DeepSeek's intelligent dialogue function: DeepSeek has virtual customer service function, which can effectively respond to multiple rounds of conversations and solve customer consultation problems

Question content

About any information in the knowledge base: DeepSeek can extract key information from large amounts of text data and generate summary, supporting intelligent decision support functions such as market trend forecasting and risk assessment.

You can choose deep thinking or concise answering methods

DeepSeek features

Smart Q&A

Ability to answer various questions asked by users, covering a wide range of knowledge areas

Text generation

You can generate high-quality articles, reports, emails and other content based on user needs

Data Analysis

Supports content extraction and analysis of uploaded files (such as text, tables, PDFs, etc.)

Multilingual support

Ability to process text in multiple languages to meet the needs of users around the world

How to use DeepSeek

Basic Operation

Enter the question you want to ask or the topic description you want to generate in the input box

Give detailed instructions, DeepSeek will generate corresponding text according to your instructions

You can continue to ask questions or request modification based on previous answers

Deep thinking (R1)

Carry out multi-angle hypothesis deduction and causal chain analysis, suitable for scenarios such as academic research, strategic analysis, and complex decision support.

For complex problems, they will be disassembled step by step to provide multi-perspective comparisons.

Search online

It can obtain the latest information on the entire network in real time, suitable for news tracking, market trends, and emergencies analysis.

Can answer dynamic updates, the latest policies of some places today's stock market, etc.

Build a personal knowledge base

Select Tool

Depending on your interests and needs, choose a DeepSeek tool to start learning

Read the document

Read the official documentation of the selected tool carefully to understand its functions, usage and precautions

Try examples

Try running and modifying the example to be familiar with the use of the tool

Advanced Learning

Understand the basics of AI knowledge such as machine learning and deep learning

Learn more about advanced features of selected tools

Practical application

Choose a project that you are interested in, such as developing a chatbot, creating a poem, etc.

Apply the learned DeepSeek to the project

Local deployment

If you consider deploying big models locally and training your knowledge base, it may be more difficult and requires more computing power.

DeepSeek Local Deployment

DeepSeek features

Support local deployment and build a personal knowledge base

You can deploy AI that is not as weak as a full-blooded version on your personal PC

Vectorized private files, such as the massive minutes of researchers, personal consumption data, medical data, etc.

DeepSeek deployment steps

Download and install the ollama tool

Install deepseek model using ollama download

Download the embedded model bge-m3

Download and install the AI visualization tool cherryStudio

Upload knowledge base files

Set ollama in cherryStudio interface, load models deepseek and bge-m3

In the cherryStudio interface, enter the dialogue interface, select the deepseek model, select the knowledge base, and ask questions

Advantages of DeepSeek

AI tools that improve production efficiency are willing to pay

Instead, lifestyles tend to be free traffic diversion

The game notebook is a computing power notebook, suitable for the local deployment of DeepSeek

illustrate

Map making tool: Mindmaster, making: empty tea cup