=Paper=
{{Paper
|id=Vol-3637/paper19
|storemode=property
|title=A Knowledge-based Service Architecture for Legal Document Building
|pdfUrl=https://ceur-ws.org/Vol-3637/paper19.pdf
|volume=Vol-3637
|authors=Silvana Castano,Alfio Ferrara,Stefano Montanelli,Sergio Picascia,Davide Riva
|dblpUrl=https://dblp.org/rec/conf/jowo/CastanoFMPR23
}}
==A Knowledge-based Service Architecture for Legal Document Building==
https://ceur-ws.org/Vol-3637/paper19.pdf
A Knowledge-Based Service Architecture for Legal
Document Building
Silvana Castano1 , Alfio Ferrara1 , Stefano Montanelli1 , Sergio Picascia1 and
Davide Riva1
1
Università degli Studi di Milano, Department of Computer Science, Via Celoria, 18 - 20133 Milano, Italy
Abstract
In this paper, we propose a knowledge-based service architecture for legal document building based on
Natural Language Processing and learning techniques, to semantically analyze a database of ingested
legal documents and propose the most prominent and pertinent textual suggestions for new document
composition. After describing the proposed NLP services for knowledge extraction and textual suggestion
selection and proposition, we describe the application of proposed document builder architecture by
considering a case study of Italian civil judgements.
Keywords
Digital Justice, Knowledge Extraction, Legal Concept Graph
1. Introduction
Legal documents constantly produced by Parliaments, Courts, and other institutional bodies
constitute a prominent source of information and knowledge not only for legal actors, like
judges or lawyers, but also for general subjects, like citizens or private and public organizations.
To improve both efficiency and effectiveness of courthouses and legal record offices, and to
foster digital justice, a significant effort is being devoted in almost all countries to digital
transformation projects, by developing legal information systems and modular architectures
providing a variety of services for acquisition, management, classification, exploration, and
retrieval of legal documents [1]. A main issue fostering digital transformation and courthouses
efficiency, is related to the availability of tools to support legal actors in the complex process
of producing a new document for the case at hand, by making available databases of previous
documentation as well as workflow management services for the different stages of the legal
proceedings [2]. In particular, advanced legal document building environments are required,
to assist legal actors in the production of a new document, like judgements or lawyer acts, by
relying on predefined document templates and knowledge-based services to properly extract
useful information available in previously produced legal documents to propose in form of
2nd Workshop on Knowledge Management and Process Mining for Law, 9th Joint Ontology Workshops (JOWO 2023),
co-located with FOIS 2023, 19-20 July, 2023, Sherbrooke, Québec, Canada
Envelope-Open silvana.castano@unimi.it (S. Castano); alfio.ferrara@unimi.it (A. Ferrara); stefano.montanelli@unimi.it
(S. Montanelli); sergio.picascia@unimi.it (S. Picascia); davide.riva1@unimi.it (D. Riva)
Orcid 0000-0002-3826-2407 (S. Castano); 0000-0002-4991-4984 (A. Ferrara); 0000-0002-6594-6644 (S. Montanelli);
0000-0001-6863-0082 (S. Picascia); 0009-0003-9681-9423 (D. Riva)
© 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
suggestions. The availability of predefined document templates [3] and rules [4] promotes a
disciplined structuring of legal documents, which will be eventually ingested in the document
repository for subsequent processing. The availability of a legal document repository where
ingested documents have a very homogeneous structure can, in turn, facilitate the application
of AI techniques, for semantic processing and classification of document contents.
In this paper, we propose a knowledge-based service architecture for legal document building
based on Natural Language Processing and Zero-Shot Learning techniques, to semantically
analyze a database of ingested legal documents and propose the most prominent and pertinent
textual suggestions for the composition of a new document. In particular, the document
building process relies on a predefined document template organized in sections, according
to a segmentation schema, and on a knowledge-extraction service and a suggestion-extraction
service to enforce the document building process. The knowledge-extraction service is responsible
for i) mining a set of featuring concepts that provide a topic-oriented description of textual
contents of ingested documents and ii) implementing a fine-grained semantic classification of
textual contents, where each concept is connected to the document portions from which the
concept emerged. The suggestion-extraction service is responsible for i) Search-by-Content, to
retrieve text portions of ingested documents that are most pertinent to the current text of the
document section under preparation; ii) Search-by-Concept, to retrieve text portions of ingested
documents that are most pertinent to one or more concepts, and possibly document section(s) of
interest, specified by the legal actor, that is, text portions that contain terminological occurrences
of the considered concept(s), coming from specified section(s), if any. The proposed service
architecture supports the legal document building process according to a “human-in-the-loop
approach”, where the document author, namely the legal actor, drafts the document under
preparation following the document template, by actively working on the text portions provided
by the suggestion service.
The paper is organized as follows. In Section 2 we introduce the proposed service architecture
for a legal document builder. Section 3 describes the knowledge-extraction service, while Section
4 illustrates the suggestion-extraction service. In Section 5 we present a case study applying the
architecture on a corpus of Italian case law decisions in the framework of the NGUPP project.
Section 6 is devoted to related work. Finally, Section 7 provides the final remarks.
2. Legal Document Builder: the Proposed Architecture
Document building, sometimes referred to as document assembly, is a process that aims at
producing a textual document following a predefined schema with the support of digital,
automated tools. The task, even when entirely performed by humans, can be thought of as
a sequence of three phases: (1) definition of the document format and structure; (2) content
draft of each part or section; (3) editing of the document. Each of these phases is susceptible to
automated support to a variable extent, ranging from interactive, human-in-the-loop approaches
to fully automated services. For instance, document structure can be totally or partially based on
predefined templates, possibly tailored to user needs. At the same time, editing may be supported
by solutions ranging from error detection to automated rephrasing and text generation.
Our proposed architecture is designed to support experts in producing documents specialized
domains that require (or would benefit from) a well-defined document structure, such as the
legal domain, which is the context we consider in this paper. A main requirement in designing
a document builder architecture for the legal domain is to preserve the importance of the
legal actor role and work with capability to provide automated support while preserving the
uniqueness/autonomy of the assessment and judgment capability of legal actors in document
writing. As a consequence, a fully automated approach is not viable nor desirable, and interactive
tool environments should be provided, to keep the legal actor “in-the-loop” and to reduce also
the risk of excessive standardization of legal documents, which is a requirement too, given the
value and purposes of legal documents.
The architecture provides a set of services to support the document assembly process by
relying on predefined document template(s). This way, document building promotes a dis-
ciplined approach to document generation, by enhancing the readability, homogeneity, and
quality of produced documentation, very important also for subsequent analysis and knowledge
extraction purposes.
The proposed service architecture for legal document building is shown in Figure 1. The
Figure 1: The proposed service architecture.
knowledge-extraction storage layer suggestion-extraction
service service
concept
graph
document legal actors
segmentation search-by-content (e.g., judges)
document chunks
ASKE and embeddings ASKE
engine engine
data
preparation ingested search-by-concept
documents
concept
(e.g., law,
mining document
judgements,
sentences) builder
legal documents environment
architecture aims to support legal actors like judges and judicial officers in the preparation
of official legal documents. In particular, the legal actor, namely the document author, starts
working by selecting a template among those available and she/he is supported in building the
document by receiving suggestions in the form of pertinent chunks of text that can be reused
“as-is” or adjusted/edited for insertion in the document under preparation. Suggestions are
extracted from a corpus of legal documents (e.g., judgements, case-law decisions) that have
been previously ingested and processed. To this end, a storage layer is defined to maintain i) the
document database for the raw ingested documents and corresponding texts; ii) the document
chunks and related embeddings extracted from documents for classification; iii) the concept
graph to store entries for the concepts extracted from documents.
Two main service pipelines are defined: the knowledge-extraction service and the suggestion-
extraction service. In the knowledge-extraction service, appropriate components are defined
to exploit the ingested documents for mining a set of featuring concepts that provide a topic-
oriented description of underlying textual contents. The concepts extracted from the documents
are organized in a graph, where a pair of similar concepts is linked by an edge. Each concept is
also connected to the document portions from which the concept emerged, meaning that we
can retrieve the pertinent document segments where a certain concept occurs. For knowledge
extraction, the involved components are document segmentation and data preparation, and
concept mining.
The suggestion-extraction service enforces two modalities to detect pertinent chunks to propose
to the user for document drafting: i) Search-by-Content, which retrieves similar chunks as
suggestions by considering the current text of the document section under preparation; ii)
Search-by-Concept, which is based on a concept specified by the legal actor to retrieve and
provide as suggestions those phrases that contain terminological occurrences of the considered
concept.
In the following sections, we describe each service in more detail.
3. The Knowledge-Extraction Service
Ingested legal documents are submitted to a knowledge-extraction pipeline based on document
annotation/segmentation, and fine-grained indexing and classification techniques.
3.1. Document Segmentation
Document segmentation is concerned with the identification of relevant sections of a document
which are exploited to construct new documents and to filter suggestions for the document
drafting. The idea is that the legal actor receives suggestions extracted from the same, or strictly
related, document section that is currently under construction.
Document segmentation is a type of document annotation that consists in dividing a document
into parts (known as segments) that “display local coherence” [5], eventually assigning a label
to each of them. Since in our case we focus only on textual data, such parts may be contiguous
sequences of words, clauses or sentences. Local coherence is here interpreted from a functional
point of view, i.e. a text segment plays a single and indivisible function in the structure of the
document, e.g. introduction, argumentation, conclusion, and others.
The first problem to solve is therefore the definition of a segmentation schema comprising
a set of functional segments and the descriptions of the respective text characteristics, which
serve as guidelines for annotation. Binding rules over the annotation output can be specified by
an additional set of axioms, including for instance a lower and/or upper bound on the number
of segments of a certain type. In most cases, definition of segment functions and characteristics
and axioms over them requires expert knowledge of the domain at hand, and shall be general
enough to encompass the variety of documents in the corpus of interest.
The second problem is the actual annotation of documents, complying with the schema.
Given a segmentation schema and a corpus to annotate, the annotation can be performed either
manually (by human annotators), automatically (by rule-based or machine learning techniques),
or by a hybrid approach.
Manual Segmentation. In the manual case, the annotation is performed by a group of human
annotators, generally, experts of the domain, who split each document into segments and
assign a label from the segmentation schema to each segment using a digital text annotation
tool. Leveraging expert knowledge can benefit annotation accuracy, especially in case of
complex documents or domain corpora like in the legal domain, but it hinders scalability,
since manual annotation is notoriously a labour-intensive, time-consuming activity. Moreover,
despite increasing accuracy, expert knowledge doesn’t ensure agreement among annotators,
which is often necessary to define a unique and homogeneous ground truth.
Automated Segmentation. Automated systems for text segmentation typically rely on either
rule-based or machine learning models. Such models ensure two main advantages with respect
to manual annotation: they can achieve higher scalability, making it possible to process large
corpora, and, if machine learning models are employed, text characteristics can be learned
automatically instead of being a-priori defined. The main disadvantage is lower accuracy,
especially on complex documents. The problem may intensify in case of totally automated
systems, which rely on pre-trained models only, in the absence of any manually annotated data.
Hybrid Segmentation. By hybrid segmentation we indicate a text segmentation system that
receives as input a limited corpus of manually annotated documents and, training and validating
an automated system on such data, extends the annotation to other documents.
In Section 5, we discuss legal document segmentation activity we performed for prototyping
a first release of a legal document builder in the framework of an ongoing research project
on digital justice in Italy. In this context, legal document segmentation has been manually
performed by legal experts of the project on a dataset case study of judgements. Since rule-based
models usually fall short in generalization, we deem machine learning models preferable to
process large and possibly heterogeneous legal corpora. We plan experimentation of different
classifiers as discussed in concluding remarks.
3.2. Data Preparation
Data preparation is concerned with the tokenization of ingested documents with the goal to
split them into document chunks. A document chunk represents the text unit to consider for
classification that can be associated with a concept. Document chunks associated with concepts
represent the suggestions that can be provided to the author while building a document. We
stress that the size of the document chunk should be large enough, so that the context can be
captured, but not too much extended to avoid segments that are long to read and potentially
noisy due to the presence of multiple concepts. In this paper, we choose to tokenize documents
by defining a chunk for few sentence/phrase detected in a document, up to a maximum size
of 128 tokens1 . This is particularly appropriate for legal actors that are typically interested in
retrieving precise suggestions in which a given concept of interest appears and can be rapidly
read/assimilated.
As a further preparation step, the terms appearing in document chunks are lemmatized and
a vector-based representation of each document chunk is finally built. The use of embedding
techniques to represent chunks allows to map the document contents on a semantic vector
space where the similarity of two chunks can be measured by comparing the corresponding
vector representations through a similarity metric (e.g., cosine similarity). For embedding
construction, Sentence-BERT [6], a modification of the original BERT model based on siamese
and triplets networks, is employed to derive a semantically meaningful embedding for a given
sentence/phrase. As such, a document chunk 𝑘 is associated with a set of terms 𝑊𝑘 therein
1
The size of the document chunk is determined by the maximum number of tokens that can be processed by the
considered embedding model.
contained. Any term is described as 𝑤 = (𝑤𝑙 , 𝑤𝑑 , 𝑤),
̄ where 𝑤𝑙 is the label of the term (i.e., the
lemma), 𝑤𝑑 is a description of the term meaning taken from a reference dictionary/vocabulary
(e.g., WordNet), and 𝑤̄ is the corresponding vector-based representation according to Sentence-
BERT, respectively. A document chunk 𝑘 has the form 𝑘 = (𝑠𝑑 , 𝑘𝑑 , 𝑘),̄ where 𝑠𝑑 is the section
of the document where the chunk occurs, 𝑘𝑑 is the original textual content of the chunk, and
𝑘̄ is the corresponding vector-based representation calculated as the mean of term vectors 𝑤̄
with 𝑤 ∈ 𝑊𝑘 . Embedding models have the capability to represent and compare the meaning of
entire text blocks like document chunks. On such a target, context-aware embedding models
fine-tuned on document similarity tasks, like Sentence-BERT, are appropriate. In the legal field,
the phrase structure can be highly articulated, and some common terms can have a precise
technical meaning when used in a court (e.g., citation, clemency, designation). Sentence-BERT
can handle such a kind of situations, which may strongly deviate with respect to everyday
conversations.
3.3. Concept Mining
Our solution to concept mining is called ASKE and it is based on zero-shot learning techniques
and context-aware embedding models to enforce concept extraction [7].
Zero-shot learning is an unsupervised classification technique, characterized by the capability
to enforce classification without requiring any pre-existing annotation of the considered docu-
ments. Initially, a seed knowledge is defined as a set of textual descriptions, each one featuring a
concept of interest, namely a seed concept, to consider for classification. Typically, for a seed
concept, a basic, gross-grained description is provided as a short text (e.g., one or two phrases)
or a list of keywords. As an example, for a seed concept about banking contract, a corresponding
textual description used for embedding is bank deposit, safe deposit box, bank credit opening, bank
advance, bank account, bank discount. Further concepts are derived from seed ones during the
extraction process, and they usually provide a more fine-grained description of the concept
instances occurring in the document chunks. A concept 𝑐, either seed or derived, is defined as a
pair 𝑐 = (𝑐𝑙 , 𝑐),̄ where 𝑐𝑙 is a label featuring the meaning of the concept expressed in a synthetic
and human-understandable way, and 𝑐 ̄ is a vector-based concept representation. Each concept 𝑐
is initially associated with the set of terms 𝑊𝑐 extracted from the textual description of 𝑐. The
vector concept 𝑐 ̄ is built as the mean of the vectors of all the terms in 𝑊𝑐 . Finally, the label 𝑐𝑙
corresponds to the label 𝑤𝑙 of the term 𝑤 ∈ 𝑊𝑐 , whose vector representation 𝑤̄ is closest to the
concept vector 𝑐.̄ Concept extraction is defined as a progressive, iterative process articulated in
the following three steps:
Zero-Shot Classification. Given a set of concepts (i.e., the seed concepts at the beginning
of the process), the document chunks are classified through zero-shot learning. A similarity
measure 𝜎, e.g. cosine similarity, is calculated over any pair of embeddings between chunks
and concepts. A document chunk 𝑘 is classified with the concept 𝑐 when the similarity value
satisfies 𝜎 (𝑘, 𝑐) ≥ 𝛼, with 𝛼 defined as a similarity threshold configured in the system. The value
of 𝛼 is empirically determined according to experimental results. In this paper, the value 𝛼 = 0.3
is employed in the proposed case-studies and experiments.
Terminology Enrichment. Given a document chunk 𝑘 classified with the concept 𝑐, the terms
in 𝑊𝑘 are exploited for enriching the term set 𝑊𝑐 . The idea is that the initial description of
the concept 𝑐 can become more detailed if we add terminology taken from chunks that are
pertinent (i.e., classified) with 𝑐. This is done by summing, for each 𝑤 ∈ 𝑊𝑘 , similarities 𝜎 (𝑤, 𝑐)
and 𝜎(𝑤, 𝐾𝑐 ), where 𝐾𝑐 denotes the average embedding of document chunks classified with 𝑐.
Terms 𝑤 ∈ 𝑊𝑘 satisfying a system-defined 𝛽 similarity threshold are inserted in 𝑊𝑐 .
Concept Derivation. By enriching the term set 𝑊𝑐 , it is possible that more fine-grained concepts
emerge from 𝑐, and they can be generated as new concepts. The discovery of possible new
concepts emerging from 𝑐 is enforced by clustering the embedding vectors 𝑤̄ of terms in 𝑊𝑐 .
The Affinity Propagation (AP) algorithm is adopted to this end, since it allows to detect the
emergence of sub-groups of similar terms within 𝑊𝑐 , without requiring to “a-priori define” the
number of clusters to generate. A new concept 𝑐 ′ is created for each cluster returned by AP
on the terms 𝑊𝑐 of a concept 𝑐. A link is defined between a concept 𝑐 ′ and 𝑐 to denote that
𝑐 ′ is derived from 𝑐 and they are somehow similar/related in content. The concept 𝑐 is then
updated since the terms in 𝑊𝑐 can be changed due to enrichment. As a consequence, 𝑐𝑙 and 𝑐 ̄ are
re-calculated.
The set of concepts obtained after derivation can trigger the execution of a new cycle based
on the above three steps. New derived concepts can contribute to improve the classification of
chunks with more fine-grained concepts. Further new concepts can be also discovered through
a new execution of enrichment and derivation on the basis of a refined classification result.
As such, concept extraction is characterized by a predefined endpoint condition based on a
termination threshold. When the number of new concepts created in the derivation step is lower
than the threshold, the concept extraction process is concluded. A concept graph providing a
topic-based description of the underlying document corpus is finally stored.
4. Suggestion-Extraction for Document Drafting
This service supports the legal document building process, by implementing a human-in-the-
loop approach, where the document author, namely the legal actor, has to draft a document
under preparation and she/he receives suggestions extracted by the ASKE engine according to
the results of the knowledge-extraction service.
The Document Builder Environment is characterized by an editing area composed of a list of
sections to be drafted, and a suggestion area where the author can retrieve pertinent chunks to
be re-used and optionally changed in the currently-edited section (see Figure 2). The author is
mainly focused on the editing area with the aim to write and create the document. The structure
of the document follows a template that the author has to choose within a predefined set. A
document template is articulated in a list of sections 𝑠1 , … , 𝑠𝑛 . In the following, we call 𝑑 ∗ the
document under editing (i.e., the active document), and 𝑠 ∗ the section on which the author is
currently focused for drafting (i.e., the active section).
Two modalities are defined to provide suggestions to the author during document drafting:
• Search-by-Content, which enforces similarity-based retrieval of text suggestions for the
section 𝑠 ∗ by exploiting the chunks of ingested documents;
• Search-by-Concept, which enforces concept-based retrieval of text suggestions for the
section 𝑠 ∗ by exploiting the concept graph.
Figure 2: The Document Builder Environment.
Editing area Suggestion area
DOCUMENT BUILDER (GUI)
choose a concept… choose a sect…
c1 s1
c2 s2
… …
SECTION s1
suggestion (chunk)
section text t1
SECTION s2 suggestion (chunk)
section text t2
SECTION s* suggestion (chunk)
suggestion (chunk)
SECTION sn
4.1. Search-by-Content
In the Search-by-Content modality, suggestions are represented by document chunks belonging
to ingested documents that are considered most similar to the active one; the degree of similarity
is computed taking into consideration document chunks belonging to a section preceding 𝑠 ∗ .
The following steps are executed to determine the most pertinent suggestions for an active
section 𝑠 ∗ with text 𝑡 ∗ .
Detection of similar documents on previous sections. Call 𝑠1 , … , 𝑠𝑘 the sections of the active
document 𝑑 ∗ that precede the active section 𝑠 ∗ . The ASKE engine is invoked to determine the
ingested documents that contain similar chunks with respect to the texts 𝑡1 , … , 𝑡𝑘 of sections
𝑠1 , … , 𝑠𝑘 , respectively. For each section 𝑠𝑖 of the active document 𝑑 ∗ with 𝑖 ≤ 𝑘, the corresponding
section content 𝑡𝑖 is passed to the ASKE engine. The ASKE embedding model, namely the model
used for knowledge extraction, is queried with the aim of extracting a vector representation ⃗𝑡𝑖 of
𝑡𝑖 2 . ASKE employs the cosine similarity function to compare ⃗𝑡𝑖 against the document chunks of
the 𝑠𝑖 section belonging to any ingested document 𝑑. For each document 𝑑, a similarity degree
𝜎𝑑 is defined by summing all the similarity values provided by the matching chunks 𝑘𝑑 in 𝑠𝑖 . For
a section 𝑠𝑖 , a set of similar documents 𝐷𝑖 is returned as a result, where the similarity degree 𝜎𝑑
of a document 𝑑 ∈ 𝐷𝑖 is over a prefixed system-designed threshold. An overall set of similar
𝑖=𝑘
documents 𝐷 = ⋂𝑖=1 𝐷𝑖 is finally returned to use as input for the next step.
Detection of pertinent chunks on the active section. Given the active section 𝑠 ∗ , we retrieve
a set of pertinent suggestions among the chunks of the similar documents in 𝐷. As a default
behavior, the chunks of section 𝑠 ∗ in the documents of 𝐷 are retrieved as suggestions to support
the author in drafting the active section. Moreover, to obtain a refined set of suggestions, the
author can start drafting the active section 𝑠 ∗ by inserting an initial section content 𝑡 ∗ . In this
case, ASKE is invoked to derive the vector representation ⃗𝑡∗ of 𝑡 ∗ . The embeddings of chunks
2
When the section content 𝑡𝑖 exceeds the size of a document chunk, a tokenization mechanism is employed to spit
𝑡𝑖 . For the sake of clarity, in the paper, we consider 𝑡𝑖 as a single textual element with a corresponding vector-based
representation ⃗𝑡𝑖 .
in documents of 𝐷 and section 𝑠 ∗ are compared against ⃗𝑡∗ . A ranked list of similar chunks are
returned as suggestions to show in the right-hand panel of Figure 2 (descending order).
4.2. Search-by-Concept
In the Search-by-Concept modality, suggestions represent the document chunks where an
occurrence of a given concept of interest appears according to the classification results of ASKE.
In the suggestion area, the author can select a concept of interest 𝑐 ∗ among a set of available
ones, namely the set of concepts derived by ASKE during knowledge extraction from the
ingested documents. The concept graph of ASKE is queried to retrieve the set of document
chunks 𝐾𝑐 ∗ classified as 𝑐 ∗ that are then returned as a result.
A further filtering option is provided in the suggestion area to enable the author to choose a
target section of interest 𝑠 ∗ . When a target section is selected, only the chunks of 𝐾𝑐 ∗ belonging
to the section 𝑠 ∗ in the ingested documents are finally shown as suggestions in the right-hand
panel of Figure 2.
5. Application to the Italian Legal Domain
The proposed service architecture has been applied to a legal case study in framework of the
Next Generation UPP Italian project, aimed at providing artificial intelligence and advanced
information management techniques for digital transformation of Italian legal processes and
digital law in general.
We firstly introduce the dataset employed in the case study, explaining document preparation
and segmentation. Then, we describe the scenario in which a legal actor, i.e. the judge, makes use
of the document builder functionalities in order to generate a new judgement, with examples of
Search-by-Content and Search-by-Concept text suggestions. The case study has been conducted
on documents written in Italian; examples reported in the paper have been translated into
English for the purpose of understanding. Also, for the sake of brevity, we report an excerpt of
the text suggestions.
5.1. Dataset
The dataset consists in a corpus of 50 Italian case law decisions, retrieved from 12 different
Courts located in Northern Italy. All the documents concern first degree civil law judgments
regarding the matter of unfair competition. Such kind of documents comes in PDF files of
different formats, depending on the court emitting it; plain text has been extracted from these
files in order to be used in the document segmentation and the data preparation processes.
For document segmentation, our legal partners of the project provided a document schema
defined on the basis of the Rules of the Court (March 2022). This set of rules has the function
of giving a rigid and predetermined structure to the introductory acts of the parties and to
the decision of the Court. In particular, for example, Rule 74 states that “any judgment of the
European Court of Human Rights must contain some basic information (such as the names of
the parties, agents, lawyers or advisers of the parties and a report of the procedure followed)
followed by some well-defined steps of the decision: the facts of the case; a summary of the
arguments of the parties; the legal reasons; operational provisions”. Similar criteria are also
established for the drafting of the appeals of lawyers. Based on Rule 74 recommendations and
related literature, our law project partners developed a document segmentation schema for
Italian juridical judgements/case law decisions [3], which has been enforced in our proposed
architecture, for document segmentation and for document building. The segmentation schema
comprises five different structural sections with the following corresponding labels (identified
directly by the domain experts): court and parties, providing information of the court, the panel
of judges and the parties in the trial; background information, relating to the proceedings of the
trial and to the reconstruction of the facts; claims and arguments, claims made by the plaintiffs
and counterclaims made by the defendants; reasoning, for each decision about an individual
claim; final decisions, for each individual claim. For the purposes of the case study, document
segmentation has been performed manually by legal domain experts involved in the project.
5.2. Search-by-Content
In the Search-by-Content scenario, the legal actor is required to provide to the system basic
knowledge regarding the case at hand, filling at least a section 𝑠 of the chosen document template,
in order to receive suggestions on an active section 𝑠 ∗ . These suggestions are retrieved from
previous relevant case law decisions stored in the database of ingested documents and sorted
based on the computed similarity scores. In particular, the legal actor could write down the
sections 𝑠1 and 𝑠2 regarding background information and claims and arguments in order to receive
suggestions for the reasoning section 𝑠 ∗ . We exclude from the analysis the first section in the
aforementioned annotation schema, court and parties, since it is auto-compiled by the software
tools currently in use by legal actors and it includes general information that end up being not
relevant for the search process.
As an example of document building, suppose to work on the generation of a new judgement
regarding the counterfeiting of machinery , for which the following two sections have
already been drafted:
Background Information: The Supreme Court confirmed what it had already ruled in previous
decisions, ruling definitively on the validity of the XXXX patent and the intervening coun-
terfeiting by YYYY; the plaintiff then filed a petition for the continuation of the suspended
trial.
Claims and Arguments: XXXX seeks relief for the contraction of revenues afferent to the
machinery, resulting from the presence of the machinery substitute of YYYY.
and the reasoning section is the active section to be generated. By invoking the Search-
by-Content modality for drafting the reasoning section, the document builder retrieves as
suggestions the most relevant document chunks of the reasoning sections of the judgements
similar to the active counterfeiting of machinery document at hand. The top-2 retrieved
document chunks ranked by similarity score are shown below.
(...) objected that such conduct challenged to XXXX would not be configurable since the
corporate purpose of the two companies is different in the sense that XXXX is limited to
carrying out only service activities for the repair of cleaning machinery, while YYYY carries
out as its prominent activity the production and sale of machinery, which XXXX does not deal
with, consequently, there could not be talk of unfair competition since the activity carried out
by the defendant company is merely marketing and not production pertaining exclusively to
the plaintiff company.
The plaintiff initially attached only pecuniary damage by referring to the royalties that
the defendants would have to pay for the use of the trademark online, then to the cost of
advertising investments for the launch of the products then not sold due to unfair competition
and alternatively the retroversion of profits. Then referring to the fact that through unfair
competition XXXX’s turnover would have unlawfully grown by increasing its holding YYYY
at the expense of ZZZZ with a reduction for the latter in the number of shares.
5.3. Search-by-Concept
In the Search-by-Concept scenario, it is possible to ask for text suggestions pertinent to one or
more concepts 𝑐 ∗ among the ones extracted by ASKE, which are representative of the actual
content of the ingested documents. Besides concept(s) of pertinence, the legal actor can also
choose the section(s) 𝑠 ∗ to which retrieved text suggestions should belong to.
Below, we report an example of retrieved text suggestions pertinent to the concept public
service located in background information section of ingested documents.
XXXX sued YYYY and ZZZZ, claiming: that it carries out funeral activities; that, by virtue of
Article 5, paragraph 2, R.L. no. 19 of 2004, ”in the event that the manager of public cemetery
and necropsy services also carries out the funeral activity referred to in Article 13 of this law,
corporate separation is mandatory...”; that, in implementation of this legislation, on March
27, 2008 the company YYYY had been established, a company that performs on behalf of the
Municipality of Rimini the cemetery and necropsy activity as provided for by Art. 1 point 3
letter c of R.L. n. 19 of 2004 i.e. funeral transport for indigent people, funeral collection and
transport on call of the Judicial Authority or for hygienic-sanitary needs, observation depot,
morgue, mortuary sanitary service, necroscopic medicine activities; by notarial deed dated
30/09/2009, YYYY conferred the commercial activity concerning funeral services to ZZZZ. (...)
The company XXXX reported: that it was dedicated to the provision of logistics and material
transport services for the ”MotoGP” world championship races and other motorcycle sports;
that it had entered into a contract with the organizing company YYYY, thus securing the
status of ”Express Supplier” and the ability to carry out all transports not only for YYYY, but
also for the various teams and suppliers participating in the MotoGP championship; that in
the structure of ZZZZ Mr. A. B. has always played a key role, following the conclusion of
a project collaboration contract, renewed until 12/31/2008, by which he had been given the
position of project manager organizing logistics for the MotoGP and other motorsport world
championships, including the contract acquisition stages, with qualifications general director
adviser and general director marketing and full decision-making and signing powers in the
name and on behalf of ZZZZ; (...)
6. Related Work
Work related to the legal document building issues regards legal document assembly, legal infor-
mation retrieval, and document segmentation. The increasing interest towards AI applications
in the legal field pushed a significant interest towards legal document assembly [2] by relying
on a predefined document structure. In [8] the authors define a clear distinction between
a document-oriented approach and an issue-oriented approach to document assembly. The
former consists in the automatic selection of document components and the instantiation of
user provided values for certain variables. The latter makes use of an explicit representation of
legal rules, whose truths of the predicates are provided by the judge in the justification. These
rules have been explicitly defined over the years in different forms, such as decision trees [9]
and interchange languages [4]. Different collection of legal documents, such as Serbian case
laws [10] and the GDPR [11], have been modeled using these approaches.
Legal information retrieval (LIR) is the discipline that aims at extracting information from a
corpus of legal documents, including case law decisions and legal codes. The digitization of
these documents produced a significant boost to the field of LIR, with many methodologies being
proposed for the problem: from boolean and rule-based approaches [12], from the exploitation
of thesauri [13] to ontologies [14]. Recent studies [15] [16] have been focused on the use of
large language models (LLM) [17], that are capable of capturing the contextual representation
of the text rather than simply focusing on the occurrences of certain terms. There have also
been studies exploiting NLP techniques directly on Italian legal documents: ontology learning
systems [18], article prediction [19], and fine-tuned LLMs [20].
The segmentation of legal documents is a task that can be performed either manually or
automatically, with the respective advantages and disadvantages. Manual segmentation [21] [22]
relies on annotations performed by field experts on a corpus that do not contain many documents
due to the complexity of the task, although the results are more precise than any automated
system. Automatic approaches to text segmentation have been developed for general corpus in
order to detect a shift in the topic discussed by subsequent sentences. In [23] semantic related
graphs, in which sentences are nodes and edges represents their relatedness, are exploited in
order to find maximal cliques. More recent studies approach the task employing global [24]
and contextual [25] vector representations of sentences. Automatic approaches have also been
developed for the segmentation of legal documents, such as US Court Decisions [26], and
Terms-of-Service documents [27].
7. Concluding Remarks
In this paper, we presented a reference service architecture for knowledge-based legal document
building. After discussing components and techniques composing the knowledge-extraction
service and the suggestion-extraction service of the proposed architecture, some examples
of the experimentation of a first legal document builder prototype on a corpus case study in
the Italian legal domain are discussed. Ongoing work is devoted to the consolidation of the
functionalities of the suggestion-extraction service of the document builder, with development
of a graphical interface for the knowledge-based document building functionalities. Future
work will be devoted to studying automated document segmentation techniques. In particular,
we envision a clause-level classification model, which first maps each clause into a vector space
exploiting a pre-trained contextual embedding model (e.g. Sentence-BERT [6]) and then applies
a classifier on the clause embedding vectors to determine the segment label of each clause. We
plan experimentation of different classifiers, operating either on single clauses, clause sequences
or clause networks.
Acknowledgments
This work is partially supported by i) the Next Generation UPP project within the PON pro-
gramme of the Italian Ministry of Justice, and ii) the project SERICS (PE00000014) under the
MUR NRRP funded by the EU - NextGenerationEU.
References
[1] M. Oswald, Algorithm-assisted decision-making in the public sector: framing the issues
using administrative law rules governing discretionary power, Philosophical Transactions
of the Royal Society A: Mathematical, Physical and Engineering Sciences 376 (2018)
20170359.
[2] T. Gordon, A theory construction approach to legal document assembly, in: Pre-
Proceedings of the Third International Conference on Logic, Informatics, and Law, Citeseer,
1989, pp. 485–498.
[3] G. Pinotti, A. Santosuosso, F. Fazio, A rule 74 for italian judges and lawyers, in: Advances
in Conceptual Modeling: ER 2022 Workshops, CMLS, EmpER, and JUSMOD, Hyderabad,
India, October 17–20, 2022, Proceedings, Springer, 2023, pp. 112–121.
[4] M. Palmirani, G. Governatori, A. Rotolo, S. Tabet, H. Boley, A. Paschke, Legalruleml:
Xml-based rules and norms., RuleML America 7018 (2011) 298–312.
[5] H. Kozima, Text segmentation based on similarity between words, CoRR cmp-lg/9601005
(1996). URL: http://arxiv.org/abs/cmp-lg/9601005.
[6] N. Reimers, I. Gurevych, Sentence-bert: Sentence embeddings using siamese bert-networks,
CoRR abs/1908.10084 (2019). URL: http://arxiv.org/abs/1908.10084. arXiv:1908.10084 .
[7] V. Bellandi, S. Castano, P. Ceravolo, E. Damiani, A. Ferrara, S. Montanelli, S. Picascia,
A. Polimeno, D. Riva, Knowledge-Based Legal Document Retrieval: A Case Study on
Italian Civil Court Decisions, in: Proc. of the 1st Int. Knowledge Management for Law
Workshop (KM4LAW), Bozen-Bolzano, Italy, 2022.
[8] L. K. Branting, An issue-oriented approach to judicial document assembly, in: Proceedings
of the 4th international conference on Artificial intelligence and law, 1993, pp. 228–235.
[9] D. B. Evans, Artificial intelligence and document assembly, Law Prac. Mgmt. 16 (1990) 18.
[10] M. Marković, S. Gostojić, Knowledge-based legal document assembly, arXiv preprint
arXiv:2009.06611 (2020).
[11] L. Robaldo, C. Bartolini, G. Lenzini, The dapreco knowledge base: representing the gdpr in
legalruleml, in: Proceedings of The 12th Language Resources and Evaluation Conference,
2020, pp. 5688–5697.
[12] W. Y. Mok, J. R. Mok, Legal machine-learning analysis: First steps towards a.i. assisted
legal research, in: Proceedings of the Seventeenth International Conference on Artificial
Intelligence and Law, ICAIL ’19, Association for Computing Machinery, New York, NY,
USA, 2019, p. 266–267. doi:10.1145/3322640.3326737 .
[13] M. C. Klein, W. Van Steenbergen, E. M. Uijttenbroek, A. R. Lodder, F. van Harmelen,
Thesaurus-based retrieval of case law., Frontiers in Artificial Intelligence and Applications
152 (2006) 61.
[14] S. Castano, A. Ferrara, M. Falduti, S. Montanelli, Crime knowledge extraction: An ontology-
driven approach for detecting abstract terms in case law decisions, in: Proc. of the 17th
Int Conference on Artificial Intelligence and Law, ICAIL ’19, ACM, New York, NY, USA,
2019, p. 179–183. doi:10.1145/3322640.3326730 .
[15] W. Hu, S. Zhao, Q. Zhao, H. Sun, X. Hu, R. Guo, Y. Li, Y. Cui, L. Ma, BERT_LF: A similar case
retrieval method based on legal facts, Wireless Communications and Mobile Computing
2022 (2022) 1–9. URL: https://doi.org/10.1155/2022/2511147. doi:10.1155/2022/2511147 .
[16] A. Ferrara, S. Picascia, D. Riva, Context-Aware Knowledge Extraction from Legal Docu-
ments through Zero-Shot Classification, in: Proc. of the 1st ER Int. Workshop on Digital
Justice, Digital Law, and Conceptual Modeling (JUSMOD22), Hyderabad, India, 2022, p. 81
– 90.
[17] J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, Bert: Pre-training of deep bidirectional
transformers for language understanding, arXiv preprint arXiv:1810.04805 (2018).
[18] A. Lenci, S. Montemagni, V. Pirrelli, G. Venturi, Ontology learning from italian legal texts,
in: Law, Ontologies and the Semantic Web, IOS Press, 2009, pp. 75–94.
[19] A. Tagarelli, A. Simeri, Unsupervised law article mining based on deep pre-trained language
representation models with application to the italian civil code, Artificial Intelligence and
Law (2021) 1–57.
[20] D. Licari, G. Comandè, Italian-legal-bert: A pre-trained transformer language model for
italian law (2022).
[21] P. Kalamkar, A. Tiwari, A. Agarwal, S. Karn, S. Gupta, V. Raghavan, A. Modi, Corpus for
automatic structuring of legal documents, arXiv preprint arXiv:2201.13125 (2022).
[22] G. Zhang, P. Nulty, D. Lillis, A decade of legal argumentation mining: Datasets and
approaches, in: Natural Language Processing and Information Systems: 27th International
Conference on Applications of Natural Language to Information Systems, NLDB 2022,
Valencia, Spain, June 15–17, 2022, Proceedings, Springer, 2022, pp. 240–252.
[23] G. Glavaš, F. Nanni, S. P. Ponzetto, Unsupervised text segmentation using semantic relat-
edness graphs, in: Proceedings of the Fifth Joint Conference on Lexical and Computational
Semantics, Association for Computational Linguistics, 2016, pp. 125–130.
[24] O. Koshorek, A. Cohen, N. Mor, M. Rotman, J. Berant, Text segmentation as a supervised
learning task, arXiv preprint arXiv:1803.09337 (2018).
[25] A. Solbiati, K. Heffernan, G. Damaskinos, S. Poddar, S. Modi, J. Cali, Unsupervised topic
segmentation of meetings with bert embeddings, arXiv preprint arXiv:2106.12978 (2021).
[26] J. Savelka, K. D. Ashley, Segmenting us court decisions into functional and issue specific
parts., in: JURIX, 2018, pp. 111–120.
[27] D. Aumiller, S. Almasian, S. Lackner, M. Gertz, Structural text segmentation of legal
documents, in: Proceedings of the Eighteenth International Conference on Artificial
Intelligence and Law, 2021, pp. 2–11.