Entity Extraction Ai | AI Final Year Project Report

Complete Technical Paper Content — Entity Extraction Ai

1. Executive Abstract, Problem Formulation & Background

Knowledge extraction is the creation of knowledge from structured (relational databases, XML) and unstructured (text, documents, images) sources. The resulting knowledge needs to be in a machine-readable and machine-interpretable format and must represent knowledge in a manner that facilitates inferencing. Although it is methodically similar to information extraction (IE) in natural language processing (NLP) and extract, transform, load (ETL), the main criterion is that the extraction result goes beyond the creation of structured information or the transformation into a relational schema. It requires either the reuse of existing formal knowledge (reusing identifiers or ontologies) or the generation of a schema based on the source data. The RDB2RDF W3C group was standardizing a language for extraction of resource description frameworks (RDF) from relational databases. Another popular example for knowledge extraction is the transformation of Wikipedia into structured data and also the mapping to existing knowledge (see DBpedia and Freebase).

Each column in the table is an attribute (i.e., predicate) Each column value is an attribute value (i.e., object) Each row key represents an entity ID (i.e., subject) Each row represents an entity instance Each row (entity instance) is represented in RDF by a collection of triples with a common subject (entity ID). So, to render an equivalent view based on RDF semantics, the basic mapping algorithm would be as follows:

create an RDFS class for each table convert all primary keys and foreign keys into IRIs assign a predicate IRI to each column assign an rdf:type predicate for each row, linking it to an RDFS class IRI corresponding to the table for each column that is neither part of a primary or foreign key, construct a triple containing the primary key IRI as the subject, the column IRI as the predicate and the column's value as the object. Early mentioning of this basic or direct mapping can be found in Tim Berners-Lee's comparison of the ER model to the RDF model.

As a preprocessing step to knowledge extraction, it can be necessary to perform linguistic annotation by one or multiple NLP tools. Individual modules in an NLP workflow normally build on tool-specific formats for input and output, but in the context of knowledge extraction, structured formats for representing linguistic annotations have been applied. Typical NLP tasks relevant to knowledge extraction include:

part-of-speech (POS) tagging lemmatization (LEMMA) or stemming (STEM) word sense disambiguation (WSD, related to semantic annotation below) named entity recognition (NER, also see IE below) syntactic parsing, often adopting syntactic dependencies (DEP) shallow syntactic parsing (CHUNK): if performance is an issue, chunking yields a fast extraction of nominal and other phrases anaphor resolution (see coreference resolution in IE below, but seen here as the task to create links between textual mentions rather than between the mention of an entity and an abstract representation of the entity) semantic role labelling (SRL, related to relation extraction; not to be confused with semantic annotation as described below) discourse parsing (relations between different sentences, rarely used in real-world applications) In NLP, such data is typically represented in TSV formats (CSV formats with TAB as separators), often referred to as CoNLL formats. For knowledge extraction workflows, RDF views on such data have been created in accordance with the following community standards:

2. Related Work, Literature Survey & Algorithmic Foundations

NLP Interchange Format (NIF, for many frequent types of annotation) Web Annotation (WA, often used for entity linking) CoNLL-RDF (for annotations originally represented in TSV formats) Other, platform-specific formats include

LAPPS Interchange Format (LIF, used in the LAPPS Grid) NLP Annotation Format (NAF, used in the NewsReader workflow management system)

Named entity recognition (NER) Coreference resolution (CO) Template element construction (TE) Template relation construction (TR) Template scenario production (ST) The task of named entity recognition is to recognize and to categorize all named entities contained in a text (assignment of a named entity to a predefined category). This works by application of grammar based methods or statistical models. Coreference resolution identifies equivalent entities, which were recognized by NER, within a text. There are two relevant kinds of equivalence relationship. The first one relates to the relationship between two different represented entities (e.g. IBM Europe and IBM) and the second one to the relationship between an entity and their anaphoric references (e.g. it and IBM). Both kinds can be recognized by coreference resolution. During template element construction the IE system identifies descriptive properties of entities, recognized by NER and CO. These properties correspond to ordinary qualities like red or big. Template relation construction identifies relations, which exist between the template elements. These relations can be of several kinds, such as works-for or located-in, with the restriction, that both domain and range correspond to entities. In the template scenario production events, which are described in the text, will be identified and structured with respect to the entities, recognized by NER and CO and relations, identified by TR.

Ontology learning is the automatic or semi-automatic creation of ontologies, including extracting the corresponding domain's terms from natural language text. As building ontologies manually is extremely labor-intensive and time-consuming, there is great motivation to automate the process.

Terminology extraction Entity linking At the terminology extraction level, lexical terms from the text are extracted. For this purpose a tokenizer determines at first the word boundaries and solves abbreviations. Afterwards terms from the text, which correspond to a concept, are extracted with the help of a domain-specific lexicon to link these at entity linking. In entity linking a link between the extracted lexical terms from the source text and the concepts from an ontology or knowledge base such as DBpedia is established. For this, candidate-concepts are detected appropriately to the several meanings of a term with the help of a lexicon. Finally, the context of the terms is analyzed to determine the most appropriate disambiguation and to assign the term to the correct concept. Note that "semantic annotation" in the context of knowledge extraction is not to be confused with semantic parsing as understood in natural language processing (also referred to as "semantic annotation"): Semantic parsing aims a complete, machine-readable representation of natural language, whereas semantic annotation in the sense of knowledge extraction tackles only a very elementary aspect of that.

3. Dataset Ingestion, Preprocessing & Stochastic Augmentations

The following table characterizes some tools for Knowledge Extraction from natural language sources.

Artificial intelligence (AI) is the capability of computational systems to perform tasks typically associated with human intelligence, such as learning, reasoning, problem-solving, perception, and decision-making. It is also a field of research spanning computer science, engineering, and mathematics that develops and studies methods and software enabling machines to perceive their environment and use learning and intelligence to take actions that maximise their chances of achieving defined goals. Prominent applications of AI include advanced web search engines, chatbots, virtual assistants, autonomous vehicles, play and analysis in strategy games (e.g., chess and Go), and content generation (e.g., text, images, audio, and videos). The traditional goals of AI research include learning, reasoning, knowledge representation, planning, natural language processing, and perception, as well as support for robotics. To reach these goals, AI researchers use techniques including state space search and mathematical optimisation, formal logic, artificial neural networks, and methods based on statistics, operations research, and economics. AI also draws upon psychology, linguistics, philosophy, neuroscience, and other fields. Some companies, such as OpenAI, Google DeepMind, and Meta, aim to create artificial general intelligence (AGI)—AI that can complete nearly any cognitive task at least as well as a human. Artificial intelligence was founded as an academic discipline in 1956. The field went through multiple cycles of optimism throughout its history, followed by periods of disappointment and loss of funding, known as AI winters. Funding and interest increased substantially after 2012, when graphics processing units (GPUs) started being used to accelerate neural networks, and deep learning outperformed previous AI techniques. This growth accelerated further after 2017 with the transformer architecture. The 2020s AI boom was spurred on largely by advances in generative AI, which can create and modify media. In addition to AI safety and unintended consequences and harms from the use of AI, ethical concerns, AI's long-term effects, environmental effects, and potential existential risks have prompted discussions of AI regulation.

AI programs use knowledge to answer questions intelligently and make deductions about real-world facts. Formal knowledge representation and knowledge engineering use symbols to represent words, concepts and things in the world. A knowledge base is a body of knowledge represented in a form that can be used by a program. An ontology is the set of objects, relations, concepts, and properties used by a particular domain of knowledge. Formal knowledge has been studied extensively since the 1970s and researchers have developed formalisms for a wide variety of difficult domains. The symbolic approach has difficulty with several problems: the breadth of commonsense knowledge (the set of atomic facts the average person knows is enormous), the sub-symbolic form of most commonsense knowledge (much of what people know is not represented as "facts" or "statements" they can express verbally), and knowledge acquisition (the problem of obtaining knowledge for AI applications). Large language models (and some other AI programs developed since 2012) do not require explicit, symbolic knowledge. They acquire knowledge by being trained on the combined text of millions of books and billions of websites. Modern AI can also learn about a domain by running experiments (as when AlphaZero learns game strategy by playing against itself). Machine learning solves the problems of general knowledge, commonsense knowledge and knowledge acquisition, however, this approach has struggled with accurate recall and valid reasoning.

Unsupervised learning analyses a stream of data, finds patterns, and makes predictions without any other guidance. Supervised learning requires labelling the training data with the expected answers, and comes in two main varieties: classification (where the program must learn to predict what category the input belongs in) and regression (where the program must deduce a numeric function based on numeric input). Reinforcement learning is when the agent is rewarded for good responses and punished for bad ones. The agent learns to choose responses that are classified as "good". Transfer learning is when the knowledge gained from one problem is applied to a new problem. Deep learning is a type of machine learning that runs inputs through biologically inspired artificial neural networks for all of these types of learning. Computational learning theory can assess learners by computational complexity, sample complexity (how much data is required), or other notions of optimisation.

Affective computing is a field that comprises systems that recognise, interpret, process, or simulate human affect (feeling, emotion, and mood). For example, some virtual assistants are programmed to speak conversationally or even banter humorously; it makes them appear more sensitive to the emotional dynamics of human interaction, or to otherwise facilitate human–computer interaction. However, this tends to give naïve users an unrealistic conception of the intelligence of existing computer agents. Moderate successes related to affective computing include textual sentiment analysis and, more recently, multimodal sentiment analysis, wherein AI classifies the effects displayed by a videotaped subject.

4. Neural Network Backbone, Layer Specifications & Loss Formulations

Local search uses mathematical optimisation to find a solution to a problem. It begins with some form of guess and refines it incrementally. Gradient descent is a type of local search that optimises a set of numerical parameters by incrementally adjusting them to minimise a loss function. Variants of gradient descent are commonly used to train neural networks, through the backpropagation algorithm. Another type of local search is evolutionary computation, which aims to iteratively improve a set of candidate solutions by "mutating" and "recombining" them, selecting only the fittest to survive each generation. Distributed search processes can coordinate via swarm intelligence algorithms. Two popular swarm algorithms used in search are particle swarm optimisation (inspired by bird flocking) and ant colony optimisation (inspired by ant trails).

Many problems in AI (including reasoning, planning, learning, perception, and robotics) require the agent to operate with incomplete or uncertain information. AI researchers have devised a number of tools to solve these problems using methods from probability theory and economics. Precise mathematical tools have been developed that analyse how an agent can make choices and plan, using decision theory, decision analysis, and information value theory. These tools include models such as Markov decision processes, dynamic decision networks, game theory, and mechanism design. Bayesian networks are a tool that can be used for reasoning (using the Bayesian inference algorithm), learning (using the expectation–maximisation algorithm), planning (using decision networks) and perception (using dynamic Bayesian networks). Probabilistic algorithms can also be used for filtering, prediction, smoothing, and finding explanations for streams of data, thus helping perception systems analyse processes that occur over time (e.g., hidden Markov models or Kalman filters).

An artificial neural network is based on a collection of nodes also known as artificial neurons, which loosely model the neurons in a biological brain. It is trained to recognise patterns; once trained, it can recognise those patterns in fresh data. There is an input, at least one hidden layer of nodes and an output. Each node applies a function and once the weight crosses its specified threshold, the data is transmitted to the next layer. A network is typically called a deep neural network if it has at least 2 hidden layers. Learning algorithms for neural networks use local search to choose the weights that will get the right output for each input during training. The most common training technique is the backpropagation algorithm. Neural networks learn to model complex relationships between inputs and outputs and find patterns in data. In theory, a neural network can learn any function. In feedforward neural networks the signal passes in only one direction. The term perceptron typically refers to a single-layer neural network. In contrast, deep learning uses many layers. Recurrent neural networks (RNNs) feed the output signal back into the input, which allows short-term memories of previous input events. Long short-term memory networks (LSTMs) are recurrent neural networks that better preserve long-term dependencies and are less sensitive to the vanishing gradient problem. Convolutional neural networks (CNNs) use layers of kernels to more efficiently process local patterns. This local processing is especially important in image processing, where the early CNN layers typically identify simple local patterns such as edges and curves, with subsequent layers detecting more complex patterns like textures, and eventually whole objects.

Deep learning uses several layers of neurons between the network's inputs and outputs. The multiple layers can progressively extract higher-level features from the raw input. For example, in image processing, lower layers may identify edges, while higher layers may identify the concepts relevant to a human such as digits, letters, or faces. Deep learning improves the performance of computer vision, speech recognition, natural language processing, image classification and many other applications.

In the late 2010s, graphics processing units (GPUs) that were increasingly designed with AI-specific enhancements and used with specialised TensorFlow software had replaced previously used central processing units (CPUs) as the dominant means for large-scale (commercial and academic) machine learning models' training. Specialised programming languages such as Prolog were used in early AI research, but general-purpose programming languages like Python have become predominant. The transistor density in integrated circuits has been observed to roughly double every 18 months—a trend known as Moore's law, named after the Intel co-founder Gordon Moore, who first identified it. Improvements in GPUs have been even faster, a trend sometimes called Huang's law, named after Nvidia co-founder and CEO Jensen Huang.

5. Training Protocol, Convergence Behavior & Experimental Evaluation

AI and machine learning technology is used in most of the essential applications of the 2020s, including:

search engines (such as Google Search) targeting online advertisements recommendation systems (offered by Netflix, YouTube or Amazon) driving internet traffic targeted advertising (AdSense, Facebook) virtual assistants (such as Siri or Alexa) autonomous vehicles (including drones, ADAS and self-driving cars) automatic language translation (Microsoft Translator, Google Translate) facial recognition (Apple's FaceID or Facebook's DeepFace and Google's FaceNet) image labelling (used by Facebook, Apple's Photos and TikTok).

AlphaFold 2 (2021) demonstrated the ability to approximate, in hours rather than months, the 3D structure of a protein. In 2023, it was reported that AI-guided drug discovery helped find a class of antibiotics capable of killing two different types of drug-resistant bacteria. In 2024, researchers used machine learning to accelerate the search for Parkinson's disease drug treatments. Their aim was to identify compounds that block the clumping, or aggregation, of alpha-synuclein (the protein that characterises Parkinson's disease). They were able to speed up the initial screening process ten-fold and reduce the cost by a thousand-fold. A 2026 Nature article titled "Dozens of AI disease-prediction models were trained on dubious data" highlighted the use of unreliable data being used to train AI medical prediction models for stroke and diabetes in 125 research articles. Evidence suggested some of the AI tools that were developed on unreliable data had been used on patients, although it was not clear if there were adverse outcomes.

Game playing programs have been used since the 1950s to demonstrate and test AI's most advanced techniques. Deep Blue became the first computer chess-playing system to beat a reigning world chess champion, Garry Kasparov, on 11 May 1997. In 2011, in a Jeopardy! quiz show exhibition match, IBM's question answering system, Watson, defeated the two greatest Jeopardy! champions, Brad Rutter and Ken Jennings, by a significant margin. In March 2016, AlphaGo won 4 out of 5 games of Go in a match with Go champion Lee Sedol, becoming the first computer Go-playing system to beat a professional Go player without handicaps. Then, in 2017, it defeated Ke Jie, who was the best Go player in the world. Other programs handle imperfect-information games, such as the poker-playing program Pluribus. DeepMind developed increasingly generalistic reinforcement learning models, such as with MuZero, which could be trained to play chess, Go, or Atari games. In 2019, DeepMind's AlphaStar achieved grandmaster level in StarCraft II, a real-time strategy game that involves incomplete knowledge of what happens on the map. In 2021, an AI agent competed in a PlayStation Gran Turismo competition, winning against four of the world's best Gran Turismo drivers using deep reinforcement learning. In 2024, Google DeepMind introduced SIMA, a type of AI capable of autonomously playing nine previously unseen open-world video games by observing screen output, as well as executing short, specific tasks in response to natural language instructions.

In January 2025, Microsoft proposed the technique rStar-Math that leverages Monte Carlo tree search and step-by-step reasoning, enabling a relatively small language model like Qwen-7B to solve 53% of the AIME 2024 and 90% of the MATH benchmark problems. Google DeepMind has developed models for solving mathematical problems: AlphaDev, AlphaEvolve, AlphaGeometry, AlphaProof, AlphaTensor, and FunSearch. When natural language is used to describe mathematical problems, converters can transform such prompts into a formal language such as Lean to define mathematical tasks. The experimental model Gemini Deep Think accepts natural language prompts directly and achieved gold medal results in the International Math Olympiad of 2025. In June 2026, an international group of mathematicians published a statement calling on mathematicians, professional organizations, and policymakers to disclose the use of AI, uphold peer review, and protect the openness and values of the discipline. In September 2026, OpenAI announced that one of their models had produced a counter-example to the Navier–Stokes existence and smoothness problem, a Millennium Prize Problem in mathematics. As of 2026, the claim has not been verified by any external mathematicians and is the subject of a priority dispute.