Article directory
TableAgent can be tried for free. After registration, you can use it 5 times for free. Once the times are used up, you can apply for certification to increase the number of times (15 times per day)
[TableAgent public beta address]
Preface
In the digital age, the importance of data analysis is as ubiquitous as the air. Business data analysis is the foundation of digital management and intelligent decision-making. At the same time, data analysis is a highly professional job. Descriptive analysis, diagnostic analysis, and predictive analysis will be daunting to most people who can only use Excel.
The TableAgent data analysis agent independently developed by Jiuzhang Yunji DataCanvas allows large models to empower personal productivity, from writing minutes and making summaries to a new level. As long as you can ask questions, you can become a senior data analyst. Discover the mysteries of data.
TableAgent feature highlights
Everyone is a data analyst
TableAgent provides privatized deployment for enterprises. The system is deployed within the enterprise and data is not leaked out, which fundamentally solves the problem of security and compliance. At the same time, TableAgent can also meet the requirements of large-scale and high-performance analysis of enterprise-level data. This is also Code Interpreter's current shortcomings.
TableAgent chose the more difficult code generation route and creatively proposed an expert model group approach to solve these problems.
Integrating new achievements in innovative applications
Another important capability brought by this release of TableAgent is professional fine-tuning. Different companies in different industries have professional language backgrounds and unique needs for analysis models in data analysis. It is difficult for general analysis tools to meet professional requirements. TableAgent provides professional fine-tuning for enterprises.
TableAgent has designed a T+ (Table Family) system for this purpose, which can efficiently implement customized fine-tuning work. At the same time, the system has the ability to self-iterate. The systematic system supports more efficient upgrades in all aspects of data analysis, allowing users to You can get an ever-upgrading data analysis experience without any awareness.
TableAgent usage introduction
After entering the platform, you can see the screenshot below, click to experience it now.
Login
You will enter the mobile phone number + verification code login page, as shown below:
Features
After successful login, you can see that the page is divided into content area and data set area. As shown below:
Users can upload relevant data set files according to their own needs. Currently, the online TableAgent version only supports data files in CSV format.
After the data set is uploaded, you can enter the question you want to initiate in the input box below.
Apply for certification
Click the Apply for Certification button in the upper right corner of the page, and the following page will open. Just fill in your personal information to submit your application for certification.
After successfully logging in, you will get 5 free usages. After successfully applying for certification, you will have 15 free usages every day within 15 days. The application page for certification is as follows:
After introducing the basic functions, let’s use it in practice together.
Actual data set analysis 1
Select the "Airbnb B&B Prices & Reviews" module in the sample data provided by the system. The CSV file is automatically provided.
Import CSV file data
The CSV file data is as follows:
Ask a question
The question I initiated is:I am planning to travel to Southeast Asia with my family. My daily budget is between 200 and 500 US dollars, and I need 3 to 6 beds. The destination is Thailand. Which is more cost-effective to go to Malaysia? Draw a picture for me to compare the cost performance.
TableAgent response result
TableAgent took 40.8 seconds for this session, using the model alaya to request and consume 3846 Tokens, and the response consumed 2734 Tokens, a total of 6580.
The answers were very detailed and quite satisfactory.
From the data we can see that the average price in Thailand is $255 and the average rating is 4.895, while the average price in Malaysia is $315.667 and the average rating is 4.97333. This could mean that prices in Thailand are relatively low but ratings are relatively high, while prices in Malaysia are relatively high but ratings are also relatively high.
However, these are averages only, and specific listing prices and ratings may vary depending on the specifics of the listing (e.g., listing type, amenities, location, etc.). Therefore, if you have specific requirements for price and ratings, it is recommended that you consider not only price and ratings, but also other factors, such as property type, facilities, location, etc., when selecting a property.
Overall, if you have high requirements on both price and ratings, Thailand may be a better choice, but if you have high requirements on other factors, Malaysia may be a better choice.
attentive service
At the end of the Q&A, TableAgent also asks next questions that may be of interest based on the current problem. In this way, users can obtain the information they want more conveniently and quickly.
TableAgent provides 3 alternative questions, namely:
- Which properties have lower discount prices but higher reviews, can I try to book them?
- Which type of property has more stable prices and can help me better control my budget?
- Which bed type has the most reviews?
Recommended questions to ask
We choose one of the questions and ask the question.
The detailed process and code are listed in the details. The core content is as follows:
Regiment:
- Calculate the price change of each type of listing: First, extract the two columns
Price(in dollar)andOffer price(in dollar)from df, and then calculate the difference between the two The absolute value of , we get the price change. Then, group by theTitlecolumn, calculate the average of each type of price movement - The housing type with the smallest price change: Find the row with the smallest
Price_Diffcolumn from df_price_diff, and take out the value of the correspondingTitlecolumn
# 首先，从df中提取'Price(in dollar)'和'Offer price(in dollar)'两列，然后计算两者的差值的绝对值，得到价格变动。然后，根据'Title'列进行分组，计算每种类型的价格变动的平均值
df['Price_Diff'] = abs(df['Price(in dollar)'] - df['Offer price(in dollar)'])
df_price_diff = df.groupby('Title')['Price_Diff'].mean().reset_index()
# 从df_price_diff中找出'Price_Diff'列最小的行，取出对应的'Title'列的值
min_price_diff_title = df_price_diff.loc[df_price_diff['Price_Diff'].idxmin(), 'Title']
According to data analysis, the housing type with the smallest price change is "{min_price_diff_title}". This means that prices for this type of property are relatively stable, making it easier for you to stay within your budget.
Actual data set analysis 2
For the second data analysis, the direction we chose is "Global Big City Population 2022-2023". Which city has the highest population growth? And draw a picture to display the top 10 with the highest growth rate. The information is as shown below:
Analyze results
According to our data analysis, the city with the highest population growth rate is Lokoja, from Nigeria, with a population growth rate of 6.11%. Next is Bujumbura, from Burundi, with a population growth rate of 5.93%. The third is Xiongan, from China, with a population growth rate of 5.85%.
The following is a bar chart of the population growth rate of these cities, which can more intuitively see the population growth of each city:
From these data, we can see that Lokoja City in Nigeria has the highest population growth rate, reaching 6.11%, followed by Bujumbura City in Burundi and Xiongan City in China, with population growth rates of 5.93% and 5.85% respectively. . The population growth rates of these four cities are very high, which may mean that the population of these cities is growing rapidly.
However, it is worth noting that these data are only analyzes based on current data, and actual population growth may be affected by many factors, such as economic conditions, policy adjustments, etc. Therefore, these data can only be used as a reference and cannot completely determine the population trend of a city.
Analyze which city has the largest future population
According to our data analysis, the city with the fastest growing population is Delhi.
We arrived at this result by calculating the population growth of each city. This is based on current population and growth rates. The specific data are as follows:
| City | Pop2023 | growthRate | |
|---|---|---|---|
| 0 | Tokyo | -78107.6184 | -0.0021 |
| 1 | Delhi | 899297.7084 | 0.0273 |
| 2 | Shanghai | 709822.6344 | 0.0243 |
| 3 | Dhaka | 754312.5200 | 0.0325 |
| 4 | São Paulo | 192267.7560 | 0.0085 |
| 5 | Mexico | City | 198304.8338 |
| 6 | Cairo | 441445.6800 0.0199 | |
| 7 | Beijing | 441854.1442 | 0.0203 |
| 8 | Mumbai | 340744.2560 | 0.0160 |
| 9 | Osaka | -45632.2416 | -0.0024 |
This table shows the projected 2023 population for each city. You can see that Delhi has the highest projected population growth rate, which means that the population of this city will grow rapidly in the coming years.
Please note that this forecast is based on current population and growth rates and is subject to change. Because population growth is affected by many factors, including but not limited to economic conditions, policies, natural disasters, etc. Therefore, this result should be viewed as a possible prediction rather than an absolute certainty.
Summarize
TableAgent, independently developed by Jiuzhang Yunji DataCanvas, provides enterprises with privatized deployment, which is safe and efficient. Based on code generation, everyone can become an advanced data analyst and easily handle descriptive, diagnostic and predictive analysis. Its specialized fine-tuning and T+ systems meet the unique needs of different industries.
Through practical data set analysis, the powerful data analysis capabilities of TableAgent are demonstrated. From Airbnb price comparisons to population growth in the world's largest cities, TableAgent delivers intuitive and detailed results. The user-friendly interface, convenient upload function and relevant recommendation question service make it stand out in the field of data analysis.
Overall, TableAgent is a powerful assistant for data analysis in the digital era, making complex data analysis simple and easy.