Patents are hard to understand and difficult to screen? Claude 3.7/DeepSeek increases patent analysis efficiency by 10 times!

Master AI-assisted patent analysis, improve efficiency, and quickly explore patent value.
Core content:
1. Three common pain points in patent analysis
2. Advantages of Claude 3.7/DeepSeek-R1 in patent analysis
3. AI prompt word template for rapid patent understanding and innovativeness evaluation
As a working professional, I know the pain of reading and analyzing patent documents, especially when I have to bear the patent indicators every year.
Obscure legal terms, lengthy and complex technical descriptions, obscure innovations...
Every time I need to research competitor patents or screen the value of my own patents, I always feel like I’m groping in a maze.
Is there any way for us to quickly understand the core content of a patent and efficiently evaluate its innovation value?
Today I will share a set of AI-assisted patent analysis methods that have been proven in actual combat, so that you can say goodbye to the pain of patent reading and increase your efficiency by at least 10 times!
Three major pain points in patent analysis
Before diving into the technology, let’s take a look at the common pain points we encounter in patent analysis:
Pain point 1: Patent documents are obscure and difficult to understand. Patent writing follows specific norms and styles and is filled with a large number of legal terms and lengthy sentences.
An ordinary patent usually has dozens or even hundreds of pages, and reading the entire patent is time-consuming and inefficient.
Pain point 2: Innovations are deliberately hidden. In order to obtain the maximum scope of protection, patent applicants often "blur" key innovations to make them difficult to directly identify, which makes it difficult for us to quickly judge the core value of a patent.
Pain point 3: Difficulty in batch screening and sorting When faced with dozens or even hundreds of patents, how can we quickly identify the most valuable and innovative patents?
Traditional manual analysis methods are almost impossible to complete efficiently.
Patented analysis capabilities of Claude 3.7 and DeepSeek-R1
Among the many large AI models, Claude 3.7 Sonnet and DeepSeek-R1 performed particularly well in patent document processing:
Claude 3.7 Sonnet has a large context window that can process the entire patent document at one time without losing key information.
It excels in structured output, logical reasoning, and multi-dimensional evaluation, and is particularly suitable for in-depth analysis and innovativeness evaluation of complex patent documents.
DeepSeek-R1 has unique advantages in processing Chinese patent documents, and is particularly good at understanding and extracting professional domain knowledge.
They have rich technical background knowledge, can accurately identify technological innovations and make reasonable assessments, and perform better in Chinese patent analysis.
Two practical prompt word templates to solve patent analysis problems
Below are two repeatedly optimized prompt word templates, which are aimed at the two core needs of "quick understanding of patents" and "patent innovation evaluation and ranking".
Template 1: Quick extraction of patent key points and generation of abstracts
This template can guide AI to extract core technical points from complex patent documents, generate structured summaries, and analyze the relevance to your own technical solutions.
As a senior expert with more than 10 years of patent examination experience in the fields of mechanics/electronics/chemistry, please follow the following specifications to assist in the writing of patent abstracts.
As the key summary of patent documents, the abstract must ensure that its technical features are complete, the expression is standardized, and the structure is rigorous.
Phase 1: Information sorting
In-depth study of the patent literature provided, focusing on identifying and refining the following elements:
-Core technical problems to be solved (clearly point out the specific defects of existing technologies)
-Key technical solutions adopted (highlighting the specific implementation methods of technical means)
- Breakthrough technological innovation points (improvements compared to existing technologies)
- Potential technology application scenarios (clarify the practical use of the technology solution)
Please summarize the above in a concise manner and avoid generalities.
Phase 2: Framework Construction
Based on the extracted core elements, construct a basic summary of 200-300 words, which should be developed in the following logical order:
-Technology classification (clearly state the technical field to which it belongs)
- Existing technical defects (objectively describe existing technical problems)
- Specific solutions (complete presentation of technical features)
-Technical advantages and highlights (elaborating on the technical effects produced)
Language requirements:
- Use standard professional terms (e.g. in the mechanical field, use "fix" instead of "fix", and use "connect" instead of "connect together")
- Keep the presentation concise and logical
- There should be a clear progressive relationship between the elements
-Use a single paragraph format and avoid paragraph breaks
- Use "~" to connect the numerical range
- Use standard punctuation
The third stage: improvement and improvement
The basic summary was systematically optimized, with a focus on the following aspects:
1. Technical content inspection:
- Completeness and accuracy of the technical features description
-The correspondence between technical means and technical effects
- Avoid introducing unnecessary implementation details
2. Language standard check:
- Eliminate repetitive content
- Remove redundant modifiers
-Avoid using words such as "the present invention"
- Eliminate marketing or promotional statements
3. Handling of special situations:
- Computer program inventions should focus on explaining the principles of solving technical problems and avoid excessive descriptions in programming language
-When mathematical formulas are involved, text descriptions should be used as the main method, and simple formulas can be cited when necessary
- When diagrams are included, they should be concise and clearly indicate key structures
Finally, please analyze the relevance of this patent to my technical solution, focusing on:
1. Similarity of technical routes
2. Relevance of problem solving
3. New ideas or inspirations that may be provided
Practical application examples
Feed the above prompt words and patents to Claude 3.7 Sonnet (I did not feed my own technical solutions here), and get the following results:
I have analyzed in detail the Chinese invention patent CN 113489585 B you provided, which is a patent on an identity authentication method for a terminal device.
## Patent Analysis Summary
This patent involves the field of smart home security, and mainly solves the technical problems in the prior art of low security protection capability and potential user information security risks when performing identity authentication on terminal devices connected to smart home devices.
### Technical Solution Overview
This patent proposes an identity authentication method based on a key algorithm, the core process of which is:
1. The terminal device sends an identity authentication request to the smart home device, and the request carries the first identity information of the terminal device
2. The smart home device generates the first public key information of the terminal device based on the first identity information, and generates the first random number ciphertext in combination with the locally generated encrypted random number
3. The smart home device returns the first random number ciphertext and the second identity information of the smart home device to the terminal device
4. The terminal device uses the first private key information to decrypt the first random number plaintext, and generates the second public key information of the smart home device based on the second identity information
5. The terminal device uses the second public key information and the encrypted random number to generate a second random number ciphertext and sends it to the smart home device
6. The smart home device uses the second private key information to decrypt the second random number ciphertext and compares the decryption result with the original random number to complete the authentication.
### Technological innovation
1. The key algorithm is used to replace the traditional username and password authentication method, which greatly improves security
2. Use identity information (which may include device identification code, user fingerprint, face information, etc.) as the basis for identity verification
3. Use random number protection during communication to effectively prevent replay attacks
4. Support multi-factor authentication, significantly improving identity authentication strength and security protection capabilities
### Technical Relevance and Inspiration
This patent has important reference value for building smart home security authentication system:
1. The idea of using asymmetric encryption algorithms (such as SM9) to replace simple password authentication is worth learning from
2. Bidirectional identity authentication mechanism enhances system security
3. Diversification of identity information (device identification code, biometrics, etc.) improves authentication reliability
4. Random number challenge-response mechanism effectively prevents replay attacks
This patent provides a systematic method for smart device identity authentication, which is of significant value in solving device access security issues in the Internet of Things environment.
The output results of Tencent Yuanbao DeepSeek-R1 full blood version:
### Phase 1: Information Sorting
**Core technical problems:**
In the prior art, smart home devices perform identity authentication through user names and passwords, which poses security risks such as brute force cracking, peeking, and replay attacks, resulting in low user information security protection capabilities.
**Key technical solutions:**
A key algorithm (such as the SM9 algorithm) is used for identity authentication, and secure communication between terminal devices and smart home devices is ensured through steps such as generating public key information, encrypting random numbers, and decrypting verification.
** Breakthrough technology innovation highlights: **
1. Use identity-based encryption algorithms to improve identity authentication strength.
2. Introduce a random number protection mechanism to prevent replay attacks.
3. Enhance security protection capabilities through multi-factor authentication.
**Potential technology application scenarios:**
Applicable to secure identity authentication between smart home devices (such as smart door locks, smart refrigerators, smart ovens, etc.) and terminal devices (such as smart phones).
### Phase 2: Framework Construction
The present invention relates to the field of smart home, and specifically to an identity authentication method for a terminal device. In the prior art, smart home devices authenticate identities through user names and passwords, which have security risks such as brute force cracking, peeking, and replay attacks, resulting in low user information security protection capabilities. In order to solve the above problems, the present invention provides an identity authentication method based on a key algorithm, including: a terminal device sends an identity authentication request to a smart home device, the smart home device generates a first public key information based on the first identity identification information of the terminal device, and generates a first random number ciphertext and returns it to the terminal device; the terminal device decrypts the first random number ciphertext, generates a second random number ciphertext and sends it to the smart home device; the smart home device decrypts the second random number ciphertext, verifies whether the random numbers are equal, and if they are equal, determines that the identity authentication is successful. The present invention improves the strength and security protection capabilities of terminal device identity authentication through an encryption algorithm based on identity identification, a random number protection mechanism, and multi-factor authentication, and solves the technical problem of user information security risks in the prior art.
### The third stage: improvement and improvement
The present invention relates to the field of smart home, and specifically to an identity authentication method for a terminal device. In the prior art, smart home devices authenticate identities through user names and passwords, which have security risks such as brute force cracking, peeking, and replay attacks, resulting in low user information security protection capabilities. In order to solve the above problems, the present invention provides an identity authentication method based on a key algorithm, including: a terminal device sends an identity authentication request to a smart home device, the smart home device generates a first public key information based on the first identity identification information of the terminal device, and generates a first random number ciphertext and returns it to the terminal device; the terminal device decrypts the first random number ciphertext, generates a second random number ciphertext and sends it to the smart home device; the smart home device decrypts the second random number ciphertext, verifies whether the random numbers are equal, and if they are equal, determines that the identity authentication is successful. The present invention improves the strength and security protection capabilities of terminal device identity authentication through an encryption algorithm based on identity identification, a random number protection mechanism, and multi-factor authentication, and solves the technical problem of user information security risks in the prior art.
### Correlation Analysis
1. **Similarity of technical routes:**
This patent and my technical solution have a high degree of similarity in technical routes, both involving secure identity authentication between smart home devices and terminal devices.
2. **Relevance of the problem solved:**
This patent proposes a solution based on a key algorithm to address the security risks of identity authentication methods in the prior art, which is highly relevant to the security issues solved in my technical solution.
3. **Possible new ideas or inspirations:**
The identity-based encryption algorithm, random number protection mechanism and multi-factor authentication introduced by this patent provide new ideas and inspiration for my technical solution, which will help to further improve security protection capabilities.
How about it? Doesn’t it look more structured and focused?
The most important thing is that they are all plain language, rather than those disgusting patent terms?
Template 2: Patent innovation analysis, scoring and ranking
This template is suitable for scenarios where you need to evaluate and sort the value of multiple patents or technical briefing documents (which can be in related or unrelated fields). It can help you quickly screen out the most valuable patents.
# Role: Patent Examiner Expert
## Background:
Senior patent examination expert with more than 10 years of experience in patent examination, especially good at patent analysis and evaluation in the fields of computer, artificial intelligence, information security, etc. Proficient in innovation point identification, patent layout and guidance on briefing document writing.
## Attention:
- This is an initial assessment of the technical solution, focusing on innovation and technical path
-Please provide the following core content:
1. Framework description of the technical solution and description of its innovations
2. 2-3 most relevant existing patents (focus on the claims)
3. Key technical features and their effects
## Constraints:
-Evaluation focuses on innovation and technical paths
- Avoid getting hung up on implementation details
-Evaluation is mainly based on the comparison files provided by users
- Suggestions should be of practical guidance
- Clearly indicate the basis for judging innovation
## Skills:
- Good at identifying technological innovations
- Proficient in creative judgment methods
-Familiar with patent layout strategy
- Ability to optimize solutions
-Professional evaluation and analysis capabilities
## Goals:
1. Analyze and evaluate the objectives:
- Identify the essential innovations of the technical solution
- Clarify the characteristics and effects of innovative technologies
-Analyze the technical contribution of innovations
-Evaluate the patent value of the solution
- Predict the possibility of authorization
2. Comparative analysis objectives:
- Comprehensive interpretation of the technical features of the comparison documents
- Extract key distinguishing features
-Analysis of technical effect differences
-Evaluate the creative level
- Identify potential infringement risks
3.Quality control objectives:
- Ensure the accuracy of innovation point analysis
- Ensure objectivity of comparative analysis
- Verify the logical integrity of the assessment
- Ensure the recommendations are actionable
-Control the professionalism of the conclusion
## Workflow:
1. Analyze innovative technical solutions
- Identify key technical features
- Extraction of innovation statement
-Understand the technical effects
2. Interpretation of comparative documents
- Analyze the claims
- Extraction of technical features
-Understand the existing technology
3. Conduct innovation assessment
- Feature comparison analysis
-Technical effect comparison
- Innovation level judgment
## OutputFormat:
Please systematically evaluate and rank the uploaded patent documents/technical briefing documents. The evaluation dimensions include technological innovation, solution integrity, legal protection, commercial value and document standardization.
Scoring results (out of 100 points):
1. Patent Name 1: XX points
-Technological innovation (25/25): [Detailed analysis of innovation points and their value]
- Solution completeness (24/25): [Detailed analysis of the solution’s logic and feasibility]
-Legal protection (18/20): [Analysis of patent protection scope and strength]
- Commercial value (15/15): [Analysis of market application prospects]
-Document Compliance (10/15): [Evaluate the quality of the document]
2. Patent Name 2: XX points
[Same format as above]
3. Patent Name 3: XX points
[Same format as above]
4. Patent Name 4: XX points
[Same format as above]
Reasons for ranking:
1. Patent 1 ranks first: [detailed reasons]
2. Patent 2 ranks second: [Detailed explanation]
3. Patent 3 ranks third: [Detailed explanation]
4. Patent 4 ranks fourth: [Detailed explanation]
This ranking is mainly based on the two most important dimensions of technological innovation and solution completeness, while also taking into account the comprehensive performance of other factors.
Remove the document standardization score and focus on technical innovation. Make sure to fully research the most relevant technical solutions and provide a detailed explanation of the technical innovation.
Practical application examples
Feeding the above prompt words and 4 patents to Claude 3.7 Sonnet, we get the following results:
# Patent Document Evaluation Report
## Rating results (total score 100 points)
### 1. "A Dynamic Authentication Risk Detection Method and System": 87 points
- **Technical Innovation (22/25)** : A dynamic authentication method based on risk assessment is proposed. By analyzing multiple authentication factors and risk prediction values, the authentication strategy is adaptively adjusted to avoid the problem that fixed authentication paths are vulnerable to attacks. The innovation lies in the introduction of a risk prediction model to dynamically decide whether to enable normal authentication mode or abnormal protection measures.
- **Completeness of the solution (24/25)** : The solution is described comprehensively, from receiving the authentication request, risk assessment, obtaining the second authentication factor to verification and risk prediction, and the logic is self-consistent. In particular, detailed steps are provided for transmission risk monitoring and the classification and processing of stable/unstable factors.
- **Legal Protectivity (16/20)** : The claims are well structured, the main claims are moderately generalized, and the dependent claims are fully developed, but the definitions of some technical features are slightly broad.
- **Commercial Value (25/30)** : With network security becoming increasingly important, this solution has high commercial value in improving identity authentication security, and is particularly suitable for areas with high security requirements such as finance and government affairs.
### 2. "Identity Authentication Method, System, Storage Medium, and Electronic Device for Terminal Devices": 83 points
- **Technical Innovation (21/25)** : A smart home device authentication mechanism based on the SM9 key algorithm and random number challenge response is proposed. The innovation lies in enhancing security through a two-way verification process of public key encryption and private key decryption, and introducing the concept of master key pairs to strengthen key management.
- **Scheme completeness (23/25)** : The scheme describes the process from key generation, exchange, authentication request to verification in a complete manner, and the design of the key generation center makes the scheme more systematic, but the description of some exception handling mechanisms is slightly insufficient.
- **Legal Protectivity (17/20)** : The claims are clearly structured, the main claims are accurately summarized, the independent and dependent claims are reasonably structured, and the scope of protection is appropriate.
- **Commercial Value (22/30)** : In the emerging field of smart home security, this solution has high application value and market potential, but the high implementation complexity may limit its application scope.
### 3. "An activity recommendation method and system based on the opening rules of smart door locks": 78 points
- **Technical Innovation (19/25)** : The method of using smart door lock opening and closing data to analyze tenant activity patterns and recommend activities. The innovation lies in the user profile created through door lock usage data without infringing on user privacy. In particular, the method of analyzing activity locations through the opening and closing status of internal and external doors is unique.
- **Completeness of solution (20/25)** : The solution is basically complete from data collection, location analysis, time statistics to similarity calculation and recommendation, but the detail and robustness of the data processing algorithm need to be strengthened.
- **Legal Protectivity (15/20)** : The structure of the claims is relatively simple, and the scope of protection of the main claims is relatively broad, but the definition of technical features is not precise enough, and some dependent claims are not fully developed.
- **Commercial Value (24/30)** : Against the backdrop of the rapid development of the long-term rental apartment market, this solution helps improve tenant experience and participation in community activities, and has good commercial prospects.
### 4. "A hotel room sales quantity prediction method and system based on big data": 81 points
- **Technical Innovation (20/25)** : A hotel room sales forecasting method based on the dual analysis of the correlation coefficient and the autocorrelation coefficient is proposed. The innovation lies in constructing a distance weight model through multi-dimensional analysis of historical occupancy data to predict future sales, and optimizing room prices in combination with dynamic game theory.
- **Completeness of solution (23/25)** : The solution has a complete process from data acquisition, calculation of cross-correlation and autocorrelation coefficients, determination of distance weights to prediction and dynamic price adjustment. The algorithm description is clear and the implementation path is clear.
- **Legal Protectivity (16/20)** : The claims are well structured and the technical features are precisely defined, but some mathematical formulas may affect the interpretation and scope of protection of the claims.
- **Commercial Value (22/30)** : It has high practical value for revenue management and resource optimization in the hotel industry, and can effectively improve hotel operating efficiency and revenue levels.
## Ranking and Reasons
1. **《A Dynamic Authentication Risk Detection Method and System》Ranked No.1** :
This patent proposes an innovative method of dynamic risk assessment in the field of security authentication. It not only has a high degree of technical integrity, but also solves the key problem that the traditional fixed authentication path is vulnerable to attack. In particular, the combination of risk prediction model and security factor forms an adaptive identity authentication mechanism, which has significant value in the context of the increasing importance of network security. This solution has both theoretical innovation and practical value, with balanced scores and the best overall performance.
2. **Terminal device identity authentication method, system, storage medium, electronic device ranked second** :
This patent proposes an authentication scheme based on asymmetric encryption and random number challenge response for smart home security, with a clear technical path and high security. Although it is slightly lower than the first-ranked patent in terms of innovation, its scheme is outstanding in terms of systematicness and integrity, especially the design of the key generation center, which makes the entire authentication system more standardized and manageable.
3. **"A Hotel Room Sales Quantity Prediction Method and System Based on Big Data" ranked third** :
This patent combines data analysis technology with dynamic game theory, and builds a prediction model through cross-correlation and autocorrelation analysis. The technical idea is novel and has strong practicality. Although it is not as innovative as the first two, the integrity of its algorithm design and its practical value in the hotel industry make it perform better than the fourth place.
4. ** "An activity recommendation method and system based on the opening rules of smart door locks" ranked fourth** :
The idea of analyzing and recommending user activities based on smart door lock data is novel, especially in terms of protecting user privacy. However, compared with the other three patents, the completeness of its technical solution and the accuracy of its algorithm still have room for improvement, and the scope of protection of the claims is not precisely defined.
## Comparative analysis of the most relevant existing technologies
### Key comparison technologies for "A dynamic authentication risk detection method and system":
1. CN 112330035 A:
-This comparison document also involves identity authentication risk control, but it mainly makes risk judgments based on static factors such as account behavior characteristics and trust environment.
-The difference lies in the different assessment objects- This application focuses on the risk assessment of the certification factors themselves, and establishes a more comprehensive risk model through stable/unstable factor classification and transmission risk monitoring
- The innovative effect of this application is that by dynamically adjusting the authentication mode, the authentication process is made adaptive, and the ability to resist different types of attacks is enhanced.
2. CN 103427999 A:
- The reference document proposes a risk-based authentication control method, but it mainly relies on preset rules and threshold judgments
-The difference is that this application introduces a method for calculating the correlation coefficient to quantitatively analyze the relationship between authentication factors and risk prediction values
-This application achieves more accurate risk assessment through risk prediction model, which can reduce the misjudgment rate and improve authentication efficiency
### Key comparison technologies for "Terminal device identity authentication method, system, storage medium, and electronic device":
1. CN 112966291 A:
-The reference document involves authentication methods for smart home devices, but is mainly based on pre-shared keys and symmetric encryption
-The difference is that this application uses the SM9 asymmetric encryption algorithm and manages key generation through the Key Generation Center (KGC), which is more secure
- This application provides two-way authentication capabilities through a random number challenge response mechanism, effectively preventing man-in-the-middle attacks and replay attacks
2. CN 111953705 A:
- This reference document proposes a device authentication method, but focuses on certificate verification and trust chain establishment
-The difference is that this application uses identity-based key generation, which does not require a complex certificate management infrastructure
- This application simplifies the key management process by introducing the concept of master key pairs while maintaining high security
### Key comparison technologies for "An activity recommendation method and system based on smart door lock opening rules":
1. CN 110232270 A:
- The reference document involves a recommendation system based on user behavior, but mainly relies on GPS location and application usage data
-The difference is that this application indirectly infers user behavior through the switch records of the smart door lock, avoiding privacy leakage
-This application is specifically targeted at long-term rental apartment scenarios. It analyzes the tenant's location by the switch status of the internal and external doors, and is targeted at specific scenarios.
2. US 8863308 B2:
- This comparative document proposes a method for analyzing user behavior patterns, but it is mainly used for security monitoring and anomaly detection
-The difference is that this application uses behavior pattern analysis for activity recommendation and considers similarity calculation in a multi-tenant environment
-This application realizes user profiling on the basis of privacy protection, balancing the needs of personalized services and privacy protection
### Key comparison technologies for "A method and system for predicting hotel room sales quantity based on big data":
1. CN 112637136 A:
- The reference document involves hotel room demand forecasting, but is mainly based on time series analysis and trend forecasting
-The difference is that this application introduces a dual analysis of the mutual correlation coefficient and the autocorrelation coefficient, taking into account the influencing factors of more dimensions
- This application combines the distance weight model and dynamic game theory, and also provides a price optimization strategy based on the prediction.
2. CN 111740828 A:
-This comparative document proposes a method for predicting accommodation reservation demand based on machine learning
-The difference is that this application does not rely on complex machine learning models, but builds a prediction model through statistical correlation analysis, achieving more explainable predictions
- This application solves the data truncation problem through the restricted dependent variable regression method and Tobit model, improving the prediction accuracy
## Key technical features and effects description
### Key technical features of "A dynamic authentication risk detection method and system":
1. **Stable/Unstable Authentication Factor Division Mechanism** :
-Compare the first authentication factor obtained with the authentication data in the database to distinguish between stable and unstable authentication factors
-Effect : Effectively identify risk points in the certification process and provide a basis for subsequent risk assessment
2. **Transmission risk monitoring mechanism** :
-Monitor the probability of normal data transmission in a specified time interval and quantify the transmission risk through mathematical models
-Effect : Timely identify network transmission risks and prevent data from being intercepted or tampered with
3. **Risk prediction model** :
- Calculate risk prediction value based on the stability/instability status and similarity assessment of the second authentication factor
-Effect : Accurately predict authentication risks and dynamically decide whether to use normal authentication or abnormal protection measures
4. **Safety factors and protection mechanisms** :
- Add security factors to the authentication process, activating the protection mechanism when the authentication errors accumulate to a threshold
-Effect : Form a double insurance mechanism to prevent brute force cracking and repeated attacks
### Key technical features of "Terminal device identity authentication method, system, storage medium, and electronic device":
1. **Key Generation Center (KGC) Architecture** :
-Deploy KGC in routers, generate master key pairs and manage key generation for terminal devices
-Effect : Achieve centralized key management, simplify key distribution process, and improve system security
2. **Identity key generation based on SM9 algorithm** :
- Generate a device-specific private key using the terminal device's identity information and master private key
-Effect : Implement identity-based encryption without complex certificate management while maintaining high security
3. **Random number challenge response authentication mechanism** :
- Devices verify each other's identity through encrypted random numbers, forming a two-way authentication process
-Effect : Effectively prevent man-in-the-middle attacks and replay attacks, ensuring communication security
4. **Key Exchange Protocol** :
- Use the Diffie-Hellman protocol to achieve shared key negotiation between the terminal device and KGC
-Effect : Establish a secure communication channel and protect the key transmission process
### Key technical features of "An activity recommendation method and system based on the opening rules of smart door locks":
1. **Inside and outside door switch status analysis** :
-Analyze the tenant's location in the room, living room or outdoors by recording the opening and closing times of internal and external doors through smart door locks
-Effect : Achieve user location awareness without violating privacy
2. **Outgoing time error correction mechanism** :
-When a tenant is detected to have left the house but there is no record of leaving the house, inferences are made based on the time when other tenants left the house
-Effect : Improve data accuracy and optimize activity location judgment
3. **Behavior similarity calculation** :
-Build a feature matrix for clustering based on the time tenants spend indoors and outdoors and the time periods they usually go out
-Effect : Identify tenant groups with similar behavior patterns, providing a basis for accurate recommendations
4. **Collaborative filtering recommendation mechanism** :
- Determine the personality characteristics of tenants based on their time in and out of the apartment, and make recommendations based on the activity records of similar tenants
-Effect : Provide personalized activity recommendations to improve user experience and community participation
### Key technical features of "A hotel room sales quantity prediction method and system based on big data":
1. **Multi-dimensional influencing factors analysis** :
- Analyze factors affecting sales from multiple perspectives, including check-in time, customer characteristics, order characteristics, and hotel characteristics
-Effect : Comprehensively consider various factors that affect room sales and improve the accuracy of the forecasting model
2. **Dual analysis of cross-correlation and autocorrelation** :
- Calculate the correlation coefficient between influencing factors and sales volume, as well as the autocorrelation coefficient between sales volume in different periods
-Effect : Analyze the impact of external factors and consider the time correlation of historical sales to form a complete analysis framework
3. **Distance Weight Model** :
-Determine the distance weight of the historical check-in date based on the cross-correlation and autocorrelation coefficients, and calculate the distance between the predicted date and the historical date
-Effect : Establish a dynamic weight prediction model to improve prediction accuracy
4. **Dynamic Game Optimization Pricing** :
- Combine sales forecasting and dynamic game theory to establish a game model between hotels and customers
-Effect : Dynamically optimize room prices and maximize hotel revenue
## Summary and Evaluation
Among the four patents evaluated, "A risk detection method and system for dynamic authentication" performed best in terms of technical innovation and solution integrity, especially its risk prediction model and security factor mechanism formed an adaptive authentication system, which effectively solved the security risks in traditional authentication methods. "Identity authentication method, system, storage medium, and electronic device for terminal equipment" provides a systematic smart home security solution based on the SM9 cryptographic algorithm and key generation center architecture, with a clear technical path and high security. "A method and system for predicting the number of hotel room sales based on big data" and "A method and system for recommending activities based on the opening rules of smart door locks" respectively demonstrated innovative applications of data analysis in specific application scenarios. Although they are not as technologically breakthrough as the first two, they have practical value in their respective fields. All four patents reflect the technical trend of data-driven decision-making and provide valuable technical solutions in their respective fields.
The output results of Tencent Yuanbao DeepSeek-R1 full blood version:
### Rating results (out of 100 points):
1. **"An activity recommendation method and system based on the opening rules of smart door locks"** : 92 points
• **Technical Innovation (25/25)** : This solution uses the opening and closing data of smart door locks to analyze tenants' activity patterns and make personalized activity recommendations. The innovation lies in using the opening data of smart door locks to avoid the privacy issues in traditional recommendation systems where users need to actively provide data. At the same time, clustering algorithms and collaborative filtering technology are used to improve the accuracy of recommendations. The technical path is clear and the innovation is significant.
• **Completeness of the solution (24/25)** : The solution has a complete logical chain from data acquisition, activity area analysis, out-of-town time period analysis, similarity calculation to activity recommendation, with clear description of technical details and strong feasibility.
• **Legal Protectivity (18/20)** : The claims cover multiple aspects such as methods and systems, and the scope of protection is relatively broad, but the technical features of some claims are described in a broad manner, which may affect the strength of protection.
• **Commercial value (15/15)** : This solution is suitable for long-term rental apartments, shared housing and other scenarios. It has clear market demand, broad application prospects and high commercial value.
2. **"A hotel room sales quantity prediction method and system based on big data"** : 88 points
• **Technical Innovation (23/25)** : This solution uses historical order data to analyze multiple factors that affect the number of room sales, and uses correlation coefficients and autocorrelation coefficients to make predictions. The innovation lies in combining multi-factor analysis with a dynamic game model to improve the accuracy of predictions. The technical path is reasonable and highly innovative.
• **Completeness of the plan (23/25)** : The plan has clear logic, detailed description of technical details, and strong feasibility, from data acquisition, calculation of correlation coefficients, calculation of autocorrelation coefficients to room sales quantity prediction.
• **Legal Protectivity (17/20)** : The claims cover multiple aspects such as methods and systems, and the scope of protection is relatively broad, but the technical features of some claims are described in a broad manner, which may affect the strength of protection.
• **Commercial Value (15/15)** : This solution is suitable for the hotel industry, has clear market demand, broad application prospects, and high commercial value.
3. **"A dynamic authentication risk detection method and system"** : 85 points
• **Technical Innovation (22/25)** : This solution dynamically adjusts the authentication mode through multi-factor authentication and risk prediction value calculation. The innovation lies in the introduction of security factors and risk level classification, which improves the security of authentication. The technical path is reasonable and highly innovative.
• **Scheme completeness (22/25)** : The scheme has clear logic, detailed description of technical details, and strong feasibility from certification request, risk assessment, risk prediction value calculation to certification mode adjustment.
• **Legal Protectivity (17/20)** : The claims cover multiple aspects such as methods and systems, and the scope of protection is relatively broad, but the technical features of some claims are described in a broad manner, which may affect the strength of protection.
• **Commercial value (14/15)** : This solution is applicable to the field of network security, has clear market demand, broad application prospects, and high commercial value.
4. **《Terminal device identity authentication method, system, storage medium, electronic device》** : 82 points
• **Technical Innovation (21/25)** : This solution uses key algorithms and random number protection to authenticate the identity of terminal devices. The innovation lies in the introduction of multi-factor authentication and random number protection, which improves the security of authentication. The technical path is reasonable and highly innovative.
• **Scheme completeness (21/25)** : The scheme has clear logic from identity authentication request, public key generation, random number ciphertext generation to identity authentication, and the technical details are described in detail, which makes it highly feasible.
• **Legal Protectivity (16/20)** : The claims cover multiple aspects such as methods and systems, and the scope of protection is relatively broad, but the technical features of some claims are described in a broad manner, which may affect the strength of protection.
• **Commercial value (14/15)** : This solution is suitable for the smart home field, has clear market demand, broad application prospects, and high commercial value.
### Ranking reason:
1. **Patent 1 ranks first** : This solution is the most outstanding in terms of technical innovation and solution completeness. It uses the door opening data of smart door locks to make personalized recommendations, avoiding the privacy issues of traditional recommendation systems. It has a clear technical path, significant innovation, clear market demand, and broad application prospects.
2. **Patent 2 ranks second** : This solution performs well in terms of technological innovation and solution completeness. It combines multi-factor analysis and dynamic game models to improve the accuracy of predictions. It has a reasonable technical path, strong innovation, clear market demand, and broad application prospects.
3. **Patent 3 ranks third** : This solution performs well in terms of technical innovation and solution integrity. It introduces security factors and risk level classification, improves the security of authentication, has a reasonable technical path, is highly innovative, has clear market demand, and has broad application prospects.
4. **Patent 4 ranks fourth** : This solution performs well in terms of technical innovation and solution integrity. It introduces multi-factor authentication and random number protection, which improves the security of authentication. It has a reasonable technical path, strong innovation, clear market demand, and broad application prospects.
This ranking is mainly based on the two most important dimensions of technological innovation and solution completeness, while also taking into account the comprehensive performance of other factors.
From the above results, we can see that the rankings given by the two models are not consistent.
Therefore, it is necessary to make subjective judgments based on the specific analysis details combined with one's own industry experience and technical accumulation.
Advanced usage tips: Let AI become your patent analysis assistant
In addition to basic patent abstract extraction and innovation assessment, these two prompt word templates can also help you achieve the following advanced functions:
1. Patent technology roadmap analysis
Upload 10-20 key patents in a certain technical field at the same time, ask AI to extract the core innovation points of each patent and analyze the technological evolution relationship between them to help you quickly sort out the technological development context in this field.
2. Patent infringement risk assessment
Upload your technical solution together with your competitor's patent documents and compare them using the innovative analysis template. AI can help you identify potential infringement risk points and provide avoidance suggestions.
3. Patent portfolio value assessment
When you need to sort out the company's patent assets, you can upload patent files in batches and let AI classify and score them according to dimensions such as technological fields and innovation value to help you identify the most valuable patent portfolio.
4. Optimization of technical briefing documents
Before applying for a patent, upload the first draft of the technical briefing document. AI can help you find out where the description is unclear and the protection is insufficient, provide professional modification suggestions, and increase the success rate of your application.
Practical tips for using Claude 3.7 and DeepSeek-R1
1. Choose a suitable model : DeepSeek-R1 is preferred for Chinese patent analysis, while Claude 3.7 Sonnet is preferred for cross-language patent analysis or more complex reasoning tasks. 2. Reasonable task decomposition : For particularly complex patents, you can first let AI extract the abstract and then conduct an innovative assessment based on the abstract, which will be more effective. 3. Provide background information : Briefly explain your technical direction or specific problem, so that AI can provide more targeted analysis. 4. Pay attention to comparative literature : If you know some relevant patents, please upload them together. AI can provide more accurate comparative analysis. 5. Confidentiality : Before uploading a patent or technical briefing document, make sure to delete or blur sensitive information.
Conclusion: The value of AI-assisted patent analysis is not just efficiency
In the fiercely competitive field of technological innovation, patent analysis capabilities have become an essential core competitiveness for R&D personnel.
With the help of advanced AI tools such as Claude 3.7 and DeepSeek-R1, not only can the efficiency of patent analysis be greatly improved, but more importantly, it can free us from the tedious literature reading and allow us to devote more energy to truly creative thinking.
Patents are no longer lofty and difficult-to-understand documents, but innovative resources that can be easily digested and quickly understood.
I hope that the methods shared in this article can help you calmly deal with patent analysis work, draw inspiration from massive patents, and push your technological innovation to a higher level.