Related judicial decision search system

Abstract

The present invention discloses a related judicial decision retrieval system, comprising: a processing unit, an input unit, an output unit and a decision database; a user inputs a factual situation data, the processing unit obtains a legal fact from the factual situation data; the processing unit obtains a legal meaning from a plurality of legal provisions; if the legal fact and the legal meaning of each legal provision contain the same essential element, the processing unit generates a search formula to search in the decision database according to the essential element, and retrieves at least one related judicial decision; the processing unit outputs the at least one related judicial decision and a statistical data through the output unit, including one or a combination of charge, degree of punishment, win rate, amount, total guilty probability, total innocent probability, partial guilty and partial innocent probability and imaginary competition result.

Description

Technical Field

[0001] This invention relates to a relevant judicial judgment retrieval system. Background Technology

[0002] When ordinary people encounter legal problems or disputes, they often need to seek help from lawyers. Based on the facts presented by the person, the lawyer searches for similar or identical practical judgments and legal opinions according to rules of experience, and provides legal advice based on these practical judgments and legal opinions, allowing the person to assess the outcome of the court's judgment on their legal problems or disputes.

[0003] To enable the public to conduct searches independently without legal assistance, Chinese Patent Publication No. CN113688635A provides a method for recommending similar cases based on semantic similarity, as follows: 1. Obtain legal documents to obtain case texts; 2. Divide the preprocessed case texts into case sample sets; 3. Train a text classification model using the case sample sets; 4. Calculate the semantic similarity of factual descriptions between two case texts in the case sample sets, thereby generating a semantic similarity sample set; 5. Train a semantic similarity calculation model using the semantic similarity sample set; 6. Obtain the query text input by the user and preprocess it; 7. Classify the preprocessed query text using the text classification model from step 3; 8. Based on the classification results from step 7, use the model generated in step 5 to calculate the similarity between the query text and case texts in the corresponding categories, and recommend several case texts with the highest similarity values. This allows for effective searching of similar cases even without legal expert knowledge.

[0004] However, judicial judgments typically use legal language to encompass the facts based on the elements of a legal provision. The general public often struggles to understand the meaning of legal terminology and is unfamiliar with the elements of a legal provision, making it difficult to accurately retrieve relevant judicial judgments. Summary of the Invention

[0005] Therefore, the purpose of this invention is to provide a relevant judicial judgment retrieval system.

[0006] This invention provides a relevant judicial judgment retrieval system for a user to search. The system includes: a processing unit for obtaining multiple legal provisions; an input unit connected to the processing unit; an output unit connected to the processing unit; and a judgment database connected to the processing unit, storing multiple judicial judgments. The user inputs factual context data into the processing unit via the input unit, and the processing unit obtains a legal fact from the factual context data. The processing unit obtains a legal meaning conforming to legal interpretation from the multiple legal provisions. The processing unit compares the legal fact with the legal meaning of each legal provision; if the legal fact and the legal meaning of one of the legal provisions contain the same element, the processing unit determines that the factual context data corresponds to... The processing unit generates a search query based on the requirement and searches the judgment database for multiple judicial judgments. When one of the judicial judgments in the judgment database contains a proportion of the requirement in the search query, and the proportion exceeds a first threshold, the processing unit considers the judicial judgment as a relevant judicial judgment and retrieves at least one relevant judicial judgment from the judgment database. The processing unit then outputs the at least one relevant judicial judgment through the output unit. The processing unit outputs statistical data obtained from the at least one relevant judicial judgment, either directly or by converting it into a chart, through the output unit. The statistical data includes one or a combination of the following: crime, sentence, success rate, amount, overall guilty probability, overall innocent probability, partial guilty and partial innocent probability, and imagined concurrent results.

[0007] This invention further provides a relevant judicial judgment retrieval system for a user to search. The system includes: a processing unit that acquires multiple legal provisions and a conversion dictionary of non-legal and legal texts; an input unit connected to the processing unit; an output unit connected to the processing unit; and a judgment database connected to the processing unit, which stores multiple judicial judgments. The user inputs factual context data in non-legal text to the processing unit via the input unit. The processing unit, based on the conversion dictionary, retrieves a legal fact in legal text from the factual context data. The processing unit obtains a legal meaning conforming to legal interpretation from the multiple legal provisions. The processing unit compares the legal fact with the legal meaning of each legal provision. If the legal fact and the legal meaning of one of the legal provisions contain the same meaning... The processing unit determines that the factual situation data corresponds to one of the legal provisions; the processing unit generates a search expression based on the requirement and searches the judgment database for multiple judicial judgments; when one of the judicial judgments in the judgment database contains a proportion of the requirement in the search expression, and the proportion exceeds a first threshold, the processing unit considers the judicial judgment as a relevant judicial judgment and retrieves at least one relevant judicial judgment from the judgment database, and the processing unit outputs the at least one relevant judicial judgment through the output unit; the processing unit outputs statistical data obtained from the at least one relevant judicial judgment directly or after converting it into a chart through the output unit, and the statistical data includes one or a combination of the crime, sentence, success rate, amount, overall guilty probability, overall innocent probability, partial guilty and partial innocent probability, and imagined concurrence results.

[0008] This invention further provides a relevant judicial judgment retrieval system for a user to search. The system includes: a processing unit that acquires multiple legal provisions and a conversion dictionary between non-legal and legal texts; an input unit connected to the processing unit; an output unit connected to the processing unit; and a judgment database connected to the processing unit, storing multiple judicial judgments. The processing unit presents a questionnaire via the output unit, the questionnaire having at least one question corresponding to at least one factual situation, and the questions using non-legal language, each question having at least one option. The user inputs factual situation data into the processing unit via the input unit by selecting the option according to the question, using non-legal language. The processing unit, based on the conversion dictionary, retrieves a legal fact in legal language from the factual situation data. The processing unit obtains a legal meaning conforming to legal interpretation from the multiple legal provisions. The processing unit compares the legal provisions with the judgments. The processing unit determines the legal context data corresponding to the legal provision if the legal fact and the legal meaning of the legal provision contain the same element. The processing unit generates a search query based on the element and searches the judgment database for multiple judicial judgments. When a judicial judgment in the judgment database contains a proportion of the element in the search query, and that proportion exceeds a first threshold, the processing unit considers that judicial judgment as a relevant judicial judgment and retrieves at least one relevant judicial judgment from the judgment database. The processing unit then outputs the at least one relevant judicial judgment via the output unit. The processing unit outputs statistical data obtained from the at least one relevant judicial judgment, either directly or by converting it into a chart, via the output unit. This statistical data includes one or a combination of the following: charge, sentence, success rate, amount, overall guilty probability, overall innocent probability, partial guilty and partial innocent probability, and imagined concurrent outcomes.

[0009] Furthermore, there are multiple types of the questionnaire, each type of which has its own set of questions; the user selects one type through the input unit, and the processing unit presents the questionnaire of that type through the output unit.

[0010] The aforementioned category includes at least one fact as defined by law.

[0011] Furthermore, the processing unit is connected to multiple sub-databases. The categories of the sub-databases are one or a combination of cause of action, main text, body text, judge's opinion, presiding judge, court of jurisdiction, relevant laws, and appointed lawyers. The processing unit decomposes each judicial judgment into multiple parts corresponding to the category of the sub-database and stores them in the sub-databases accordingly. Before the processing unit searches the judgment database for the judicial judgment using the search formula, the processing unit searches the sub-databases using the search formula to obtain the parts related to the search formula. If the repetition between the relevant laws of the judicial judgments to which the retrieved parts belong is less than a second threshold, or if they are all different from one of the legal provisions, the processing unit regenerates the search formula according to the requirement.

[0012] Furthermore, after obtaining the at least one relevant judicial judgment, the processing unit outputs a lawyer's legal analysis opinion corresponding to the at least one relevant judicial judgment through the output unit. The lawyer's legal analysis opinion is made based on the judge's opinion on the at least one relevant judicial judgment.

[0013] The factual contextual data includes behavioral data and optionally includes additional data and / or personalized data.

[0014] The processing unit is signal-connected to a public database that stores multiple historical judgment files. The processing unit obtains the historical judgment files from the public database and performs text recognition and extraction on the historical judgment files to construct and / or update the judgment database.

[0015] The conversion dictionary is a table comparing the same legal fact in non-legal and legal text.

[0016] Based on the above technical features, the following effects can be preferably achieved:

[0017] 1. Even if users do not have legal expertise or knowledge of legal texts, they can still retrieve the most relevant judicial decisions through the processing unit's conversion.

[0018] 2. Further adopting a questionnaire format without legal text structure, users can more intuitively select the appropriate option based on the legal facts they encounter, and more easily input factual situational data.

[0019] 3. Before answering the questions in the questionnaire, select the type of questionnaire and only display the selected type of questionnaire. This can reduce the number of questions required for the same questionnaire, thereby saving users time in browsing the questionnaire.

[0020] 4. By searching the sub-database with the search query, the search query can be verified, thereby improving the accuracy of subsequent searches for relevant judicial judgments.

[0021] 5. After obtaining the relevant judgment, the system will output legal analysis opinions from lawyers and / or statistical data to help users understand the reasons and trends of the relevant judicial judgments, and thus predict the possible judicial judgment results they may face in their own legal situations.

[0022] 6. In addition to user behavior data, factual contextual data can also include important data and / or personalized data to improve the accuracy of subsequent retrieval of relevant judicial decisions.

[0023] 7. The judicial judgments in the judgment database are compiled from historical judgment archives in public databases and can be updated regularly or at any time. Attached Figure Description

[0024] Figure 1 This is a system block diagram of an embodiment of the present invention.

[0025] Figure 2 This is a functional block diagram of an embodiment of the present invention.

[0026] Figure 3 This is a flowchart of an embodiment of the present invention.

[0027] Figure 4 This is a schematic diagram of an embodiment of the present invention. Figure 1 This indicates the selection of different types of questionnaires.

[0028] Figure 5 This is a schematic diagram of an embodiment of the present invention. Figure 2 This indicates that you can fill out the questionnaire by selecting options.

[0029] Figure 6 This is a schematic diagram of an embodiment of the present invention. Figure 3 This indicates that you are entering personalized data.

[0030] Figure 7 This is a schematic diagram of an embodiment of the present invention. Figure 4 This indicates the display of statistical data and related judicial rulings.

[0031] Figure 8 This is a schematic diagram of an embodiment of the present invention. Figure 5 This indicates the display of statistical data and legal analysis opinions from lawyers.

[0032] Figure labeling: 1-Processing unit; 2-Input unit; 3-Output unit; 4-Judgment database; 41-Judicial judgment; 5-Sub-database; A-Public database; A1-Historical judgment archives; B-Factual situational data; B'-Legal facts; C0-Type; C-Questionnaire; C1-Questions; C11-Options; D-Legal provisions; D'-Legal significance; E-Search query; F-Related judicial judgments; G-Statistical data; H-Lawyer's legal analysis opinion. Detailed Implementation

[0033] Based on the above technical features, the main functions of the judicial judgment retrieval system of the present invention will be clearly presented in the following embodiments.

[0034] Please see Figure 1 and Figure 2 The invention discloses a judicial judgment retrieval system that can be used as a program product or a computer-readable medium storing the program product, so that after being loaded into a computer, a user can operate it to execute a judicial judgment result prediction method or a judicial judgment retrieval method.

[0035] The relevant judicial judgment retrieval system includes: a processing unit 1, an input unit 2, an output unit 3, a judgment database 4, and multiple sub-databases 5. The input unit 2, the output unit 3, the judgment database 4, and the sub-databases 5 are all signal-connected to the processing unit 1. The processing unit 1 is also signal-connected to a public database A.

[0036] Preferably, the processing unit 1, the input unit 2, and the output unit 3 can be integrated into the computer, and connected to the decision database 4 and the sub-database 5 in the cloud or in a physical location via the computer. For example, the input unit 2 may be a screen, and the output unit 3 may be a keyboard or mouse, or the input unit 2 and the output unit 3 may be integrated into a touch screen, etc. The present invention is not limited thereto.

[0037] It should be noted that the preferred embodiment of the present invention is described using only a single processing unit 1. In actual implementation, each database (such as the decision database 4 and the sub-database 5) may have its own processing unit 1, or the databases may perform operations in a relational database mode, etc. The present invention does not impose any limitations.

[0038] The judgment database 4 stores multiple judicial judgments 41, which are compiled from multiple historical judgment files A1 stored in the public database A. The judgment database 4 can be updated regularly or at any time.

[0039] More specifically, the processing unit 1 obtains the historical judgment file A1 from the public database A, and the processing unit 1 performs text recognition and extraction on the historical judgment file A1 using Optical Character Recognition (OCR) technology to construct and / or update the judgment database 4.

[0040] The categories of the sub-database 5 are: cause of action, main text, body text, judge's opinion, presiding judge, court of jurisdiction, relevant laws, and appointed lawyer, or a combination thereof.

[0041] The processing unit 1 will decompose each judicial judgment 41 into multiple parts corresponding to the category of the sub-database 5 by means of paragraph headings or by setting start and end characters, and store them in the sub-database 5 accordingly.

[0042] When a user sets up the relevant judicial judgment retrieval system and wants to execute the prediction method or the judicial judgment retrieval method, they can input factual situation data B in non-legal text into the processing unit 1 through the input unit 2.

[0043] For example, the processing unit 1 can present a questionnaire C for each of a plurality of types C0 via the output unit 3 [the types C0 and the questionnaire C are respectively matched]. Figure 4 and Figure 5 The user is asked to fill in the form of questions and answers. The category C0 includes at least one fact as defined by any law. It can be a single fact or a combination of multiple facts, such as fraud, murder, assault, etc. This is only an example and is not a limitation.

[0044] The processing unit 1 also acquires multiple legal provisions D and a conversion dictionary between non-legal and legal texts. Legal provisions D can also originate from the public database A or legal data on the Internet. The conversion dictionary can be imported, for example, via the cloud or a physical hard drive. This conversion dictionary is a table comparing non-legal and legal texts, especially for the same legal fact B'. Legal writing refers to the language used when writing or explaining content related to law, including professional terminology. Non-legal writing refers to commonly used language that is not legal writing, or vernacular language that is generally understandable to the average person.

[0045] The processing unit 1, based on the conversion dictionary, filters the text of the factual situation data B (non-legal text) by frequency of occurrence, selects the most frequently occurring keywords or keywords that exist in the conversion dictionary, and then decomposes the keywords and converts them into legal text. Alternatively, it first converts the factual situation data B (non-legal text) into legal text, then selects keywords, and then obtains the keywords of legal text from the factual situation data B as the legal fact B'.

[0046] The factual context data B includes behavioral data and optionally includes additional data and / or personalized data. For example, the behavioral data could be acts such as forgery or theft, and could correspond to one or a combination of behaviors. The additional data could be the user's status as a civil servant, whether the other party is an adult, etc., and the personalized data could be the court in the user's location, the judge presiding over the user's case, etc.

[0047] The processing unit 1 also filters the text of the legal provision D by frequency of occurrence to select the most frequently occurring keywords or those that exist in the conversion dictionary, obtains the keywords of the legal text from the legal provision D as multiple conditions for the legal provision D, and then combines at least one condition to form a legal meaning D' that conforms to legal interpretation.

[0048] The processing unit 1 compares the legal fact B' with the legal meaning D' of each legal provision D. If the legal fact B' and the legal meaning D' of one of the legal provisions D contain the same element, the processing unit 1 determines that the factual situation data B corresponds to one of the legal provisions D.

[0049] Since there may be more than one element, for example, the elements of drunk driving may include driving a vehicle, drinking alcohol, etc., the processing unit 1 then generates a search expression E in Boolean form based on the element, and searches for the judicial judgment 41 in the judgment database 4.

[0050] When one of the judicial judgments 41 in the judgment database 4 contains a proportion of the elements in the search query E, and the proportion exceeds a first threshold, for example, when the number of elements is 10, and one of the judicial judgments 41 contains 9 of the elements, exceeding the first threshold set to 85%, the processing unit 1 will regard one of the judicial judgments 41 as a related judicial judgment F, and retrieve at least one related judicial judgment F from the judgment database 4, and the processing unit 1 will output the at least one related judicial judgment F through the output unit 3.

[0051] Please see Figures 1 to 3In a preferred embodiment of the present invention, before the processing unit 1 searches for the judicial judgment 41 in the judgment database 4 using the search formula E, the processing unit 1 first searches for the sub-database 5 using the search formula E to obtain the part related to the search formula E.

[0052] If the repetition between the relevant laws of the judicial judgments 41 to which the retrieved part belongs is less than a second threshold, or if they are all different from one of the legal provisions D, the search formula E may be inappropriate, and the processing unit 1 will regenerate the search formula E according to the requirement.

[0053] For example, if the first judicial judgment 41 retrieved applies to 3 laws and the second judicial judgment 41 applies to 2 laws, and there is only 1 law that is common between the first judicial judgment 41 and the second judicial judgment 41, then the repetition between the first judicial judgment 41 and the second judicial judgment 41 can be calculated as 25%, which is lower than the second threshold of 50%. The processing unit 1 will then adjust the search expression E, for example, by changing the original OR Boolean operation relationship to AND, etc.

[0054] After repeated comparisons and adjustments, and once the retrieval formula E is verified to be appropriate, the processing unit 1 uses the retrieval formula E to search for the judicial judgment 41 in the judgment database 4, which can improve the retrieval accuracy of the relevant judicial judgment F by the retrieval formula E.

[0055] Please see Figure 1 , Figure 2 , Figure 4 and Figure 5 The following example uses "fraud cases" to illustrate this point.

[0056] When the user inputs the factual situation data B into the processing unit 1, the processing unit 1 presents the questionnaire C for each of the types C0 via the output unit 3.

[0057] Since it is a fraud case, the user can select the questionnaire C of fraud type C0 through the drop-down menu or by clicking on the input unit 2. The processing unit 1 will then present only the questionnaire C of fraud type C0 through the output unit 3 for the user to fill out.

[0058] By selecting the type C0, the number of questions required for the same questionnaire C can be reduced, thereby saving the user time in browsing the questionnaire C.

[0059] Each type of questionnaire C0 has multiple questions C1 corresponding to at least one factual situation, which corresponds to the legal fact B', and each question C1 has at least one option C11. Preferably, the wording of the questions C1 does not use legal terminology.

[0060] The user inputs the factual situation data B through the input unit 2 by selecting option C11 according to the question C1.

[0061] For example, if the user has forged official documents and forged securities, they can select option C11 for "Intentionally creating or altering documents" and "Intentionally creating securities".

[0062] In a preferred embodiment of the present invention, the behavioral data and the important data are mixed as option C11 of question C1. In actual implementation, the behavioral data and the important data can also be separated to avoid the user from omitting information and to improve the accuracy of subsequent retrieval of relevant judicial judgments F.

[0063] Please see Figure 1 , Figure 2 and Figure 6 The user can continue to input personalized data. For example, if they want to view all courts, they can select all courts through a drop-down menu or by clicking. They can also input the name of the presiding judge, the name of the appointed judge, and so on, as needed.

[0064] Next, the processing unit 1 performs the retrieval. The detailed retrieval process is as described above and will not be repeated here.

[0065] Please see Figure 1 , Figure 2 and Figure 7 After obtaining the relevant judicial judgment F, the processing unit 1 directly or converts the statistical data G obtained from the relevant judicial judgment F into a chart and outputs it through the output unit 3 as a judicial judgment result predicted by the processing unit 1.

[0066] The processing unit 1 can also output the relevant judicial judgment F via the output unit 3, and arrange them in a manner such as relevance and time sequence.

[0067] The statistical data G can be, for example, one or a combination of the following: crime, sentence, success rate, amount involved, probability of total guilt, probability of total innocence, probability of partial guilt and partial innocence, and imagined concurrence of outcomes. For example, the crime is a name extracted from the information of the cause of action that conforms to the legal definition, such as fraud, or the keyword "crime" appearing in the main text, etc. The other statistical data G can be deduced in the same way.

[0068] Please see Figure 1 , Figure 2 and Figure 8 ,Apart from Figure 7 The processing unit 1 can also output the legal analysis opinion H of a lawyer corresponding to the relevant judicial judgment F via the output unit 3.

[0069] The lawyer's legal analysis opinion H is based on the judge's reasoning regarding the relevant judicial judgment F. It can be an opinion drafted by the lawyer and stored in the sub-database 5. Preferably, when the user inputs the same or similar factual situation data B, the processing unit 1 can directly retrieve the stored lawyer's legal analysis opinion H from the sub-database 5 and output it directly by the output unit 3, without further drafting by the lawyer. The lawyer's legal analysis opinion H concisely explains the reasons and justifications for the judge's judgment based on the lawyer's legal reasoning and empirical rules, such as the legal fact B', post-offense attitude, family situation, etc., allowing the user to quickly understand the relevant judicial judgment F and even seek opportunities for reduced sentences or fines based on it. For example, assuming the judge of the relevant judicial judgment F might reduce the defendant's sentence due to a good post-offense attitude, the user can adopt litigation strategies such as cooperating with the investigation, making proactive statements, or even plea bargaining to strive for a more favorable judgment outcome.

[0070] Please refer to the following: Figure 1 , Figure 2 and Figure 5 Even if the user lacks legal expertise or knowledge of legal terminology, they can still intuitively select the appropriate option C11 based on their situation, without needing to manually consult and input search keywords, thus more easily inputting the factual situation data B. Then, through the transformation process of the processing unit 1, the most relevant judicial judgment F can be retrieved.

[0071] Please see Figure 7 and Figure 8 The user can use the statistical data G and the lawyer's legal analysis H to estimate the possible judicial outcome of the legal situation they encounter.

[0072] If the statistical result obtained by the user using the judicial judgment retrieval method is that the probability of all acquittals is 99%, but the actual judgment result is all guilty, the user can determine, either by himself or by entrusting a lawyer, whether the judge actually followed the rules of experience, the rules of reasoning, and the rules of evidence in the actual judgment, and thus strive to protect his own rights.

[0073] Based on the above description of the embodiments, one can fully understand the operation, use, and effects of the present invention. However, the above embodiments are only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Simple equivalent changes and modifications made in accordance with the claims and description of the present invention are all within the scope of the present invention.

Claims

1. A relevant judicial judgment retrieval system for use by a user, characterized in that, The relevant judicial judgment retrieval system includes: One processing unit obtains a complex number of legal provisions; An input unit, the signal is connected to this processing unit; An output unit, with signals connected to the processing unit; and A judgment database is signal-connected to the processing unit, and the judgment database stores multiple judicial judgments; the user inputs factual situation data into the processing unit via the input unit, and the processing unit obtains a legal fact from the factual situation data; This processing unit derives a legal meaning from the multiple legal provisions, conforming to legal interpretation. The processing unit compares the legal fact with the legal meaning of each legal provision. If the legal fact and the legal meaning of one of the legal provisions contain the same element, the processing unit determines that the factual situation data corresponds to one of the legal provisions. The processing unit generates a search query based on the requirement and searches the judgment database for the plurality of judicial judgments. The processing unit is connected to a plurality of sub-databases. The categories of the plurality of sub-databases are one or a combination of cause of action, main text, body text, judge's opinion, presiding judge, court of jurisdiction, relevant laws and appointed lawyers. The processing unit decomposes each judicial judgment into a plurality of categories corresponding to the sub-databases and stores them in the sub-databases accordingly. Before the processing unit searches the judgment database for the judicial judgment using the search formula, the processing unit searches the sub-database using the search formula to obtain the part related to the search formula. If the repetition between the relevant laws of the judicial judgments to which the retrieved part belongs is less than a second threshold, or if they are all different from one of the legal provisions, the processing unit adjusts the Boolean operation relationship in the search formula from "OR" to "AND", and regenerates the search formula, repeating the trial search until the repetition between the relevant laws is not less than the second threshold. Then, the processing unit uses the adjusted search formula to conduct a formal search for the judicial judgment in the judgment database. When one of the judicial judgments in the judgment database contains a proportion of the elements in the search formula, and the proportion exceeds a first threshold, the processing unit regards the judicial judgment as a relevant judicial judgment, retrieves at least one relevant judicial judgment from the judgment database, and outputs the at least one relevant judicial judgment through the output unit. The processing unit outputs statistical data obtained from the at least one relevant judicial judgment, either directly or after converting it into a chart, through the output unit. The statistical data includes one or a combination of the following: crime, sentence, success rate, amount, overall guilty probability, overall innocent probability, partial guilty and partial innocent probability, and imagined concurrence of events.

2. The relevant judicial judgment retrieval system as described in claim 1, characterized in that, Furthermore, after obtaining the at least one relevant judicial judgment, the processing unit outputs a lawyer's legal analysis opinion corresponding to the at least one relevant judicial judgment through the output unit. The lawyer's legal analysis opinion is made based on the judge's opinion on the at least one relevant judicial judgment.

3. The relevant judicial judgment retrieval system as described in claim 1, characterized in that, The factual contextual data includes behavioral data and optionally includes additional data and / or personalized data.

4. The relevant judicial judgment retrieval system as described in claim 1, characterized in that, The processing unit is signal-connected to a public database that stores multiple historical judgment files. The processing unit obtains the historical judgment files from the public database and performs text recognition and extraction on the historical judgment files to construct and / or update the judgment database.

5. A relevant judicial judgment retrieval system for use by a user, characterized in that, The relevant judicial judgment retrieval system includes: One processing unit obtains a dictionary of multiple legal provisions and a conversion dictionary between non-legal and legal texts; An input unit, the signal is connected to this processing unit; An output unit, with signals connected to the processing unit; and A judgment database is signal-connected to the processing unit, and the judgment database stores multiple judicial judgments; the user inputs factual situation data in non-legal text to the processing unit via the input unit, and the processing unit obtains a legal fact in legal text from the factual situation data according to the conversion dictionary; This processing unit derives a legal meaning from the multiple legal provisions, conforming to legal interpretation. The processing unit compares the legal fact with the legal meaning of each legal provision. If the legal fact and the legal meaning of one of the legal provisions contain the same element, the processing unit determines that the factual situation data corresponds to one of the legal provisions. The processing unit generates a search query based on the requirement and searches the judgment database for the plurality of judicial judgments. When one of the judicial judgments in the judgment database contains a proportion of the elements in the search formula, and the proportion exceeds a first threshold, the processing unit regards the judicial judgment as a relevant judicial judgment, retrieves at least one relevant judicial judgment from the judgment database, and outputs the at least one relevant judicial judgment through the output unit. The processing unit outputs statistical data obtained from the at least one relevant judicial judgment, either directly or after converting it into a chart, through the output unit. The statistical data includes one or a combination of the following: crime, sentence, success rate, amount, overall guilty probability, overall innocent probability, partial guilty and partial innocent probability, and imagined concurrence of events.

6. The relevant judicial judgment retrieval system as described in claim 5, characterized in that, This conversion dictionary is a table comparing the same legal facts in non-legal and legal text.

7. A relevant judicial judgment retrieval system for use by a user, characterized in that, The relevant judicial judgment retrieval system includes: A processing unit obtains a dictionary of multiple legal provisions and a conversion dictionary between non-legal and legal texts; An input unit, the signal is connected to this processing unit; An output unit, with signals connected to the processing unit; and A judgment database, signal-connected to the processing unit, stores multiple judicial judgments; The processing unit presents a questionnaire via the output unit. The questionnaire has at least one question corresponding to at least one factual situation, and the language used in the questions is not legal. Each question has at least one option. The user inputs factual situation data in non-legal text to the processing unit by selecting at least one of the options according to at least one of the questions through the input unit. The processing unit then obtains a legal fact in legal text from the factual situation data according to the conversion dictionary. This processing unit derives a legal meaning from the multiple legal provisions, conforming to legal interpretation. The processing unit compares the legal fact with the legal meaning of each legal provision. If the legal fact and the legal meaning of one of the legal provisions contain the same element, the processing unit determines that the factual situation data corresponds to one of the legal provisions. The processing unit generates a search query based on the requirement and searches the judgment database for the plurality of judicial judgments. When one of the judicial judgments in the judgment database contains a proportion of the elements in the search formula, and the proportion exceeds a first threshold, the processing unit regards the judicial judgment as a relevant judicial judgment, retrieves at least one relevant judicial judgment from the judgment database, and outputs the at least one relevant judicial judgment through the output unit. The processing unit outputs statistical data obtained from the at least one relevant judicial judgment, either directly or after converting it into a chart, through the output unit. The statistical data includes one or a combination of the following: crime, sentence, success rate, amount, overall guilty probability, overall innocent probability, partial guilty and partial innocent probability, and imagined concurrence of events.

8. The relevant judicial judgment retrieval system as described in claim 7, characterized in that, Furthermore, there are multiple types of the questionnaire, each type of which has its own set of questions; the user selects one type through the input unit, and the processing unit presents the questionnaire of that type through the output unit.

9. The relevant judicial judgment retrieval system as described in claim 8, characterized in that, The category includes at least one fact as defined by law.

Images