Quantum computing has shown promise for solving complex optimization problems in databases, such as join ordering and index selection. Prior work often submits formulated problems directly to black-box quantum or quantum-inspired solvers with the expectation of directly obtaining a good final solution. Due to the black-box nature of these solvers, users cannot perform fine-grained control over the solving procedure to balance the accuracy and efficiency, which in turn limits flexibility in real-time settings where most database problems arise. Moreover, it leads to limited potential for handling large-scale database optimization problems. In this paper, we propose a vision for the first real-time quantum-augmented database system, enabling transparent solutions for database optimization problems. We develop two complementary scalability strategies to address large-scale challenges, overcomplexity, and oversizing that exceed hardware limits. We integrate our approach with a database query optimizer as a preliminary prototype, evaluating on real-world workload, achieving up to 14x improvement over the classical query optimizer. We also achieve both better efficiency and solution quality than a black-box quantum solver.

Database systems encompass several performance-critical optimization tasks, such as join ordering and index tuning. As data volumes grow and workloads become more complex, these problems have become exponentially harder to solve efficiently. Quantum computing, especially quantum annealing, is a promising paradigm that can efficiently explore very large search spaces through quantum tunneling. It can escape local optima by tunneling through energy barriers rather than climbing over them. Earlier works mainly focused on providing an abstract representation (e.g., Quadratic Unconstrained Binary Optimization (QUBO)) for the database optimization problems (e.g., join order) and overlooked the real integration within database systems due to the high overhead of quantum computing services (e.g., a minimum 5s runtime for D-Wave's CQM-Solver). Recently, quantum annealing providers have offered more low-latency solutions, e.g., NL-Solver, which paves the road to actually realizing quantum solutions within DBMSs. However, this raises new systems research challenges in balancing efficiency and solution quality. In this talk, we show that this balance is possible to achieve. As a proof of concept, we present Q2O, the first real Quantum-augmented Query Optimizer. We show the end-to-end workflow: we encode the join order problem as a nonlinear model, a format solvable by the NL-Solver, using actual database statistics; the solution is translated into a plan hint that guides PostgreSQL's optimizer to produce a complete plan. Q2O is capable of handling actual queries in real time.

Quantum computing has shown significant potential to address complex optimization problems; however, its application remains confined to specific problems at limited scales. Spatial regionalization remains largely unexplored in quantum computing due to its complexity and large number of variables. In this paper, we introduce the first hybrid quantum-classical method to spatial regionalization by decomposing the problem into manageable sub-problems, leveraging the strengths of both classical and quantum computation. This study establishes a foundational framework for effectively integrating quantum computing methods into realistic and complex spatial optimization tasks. Our initial results show a promising quantum performance advantage for a broad range of spatial regionalization problems and their variants.

Query optimization is a crucial problem in database systems that has been studied for decades. Learned query optimizers (LQOs) can improve performance over time by incorporating feedback; however, they suffer from cold-start issues and often require retraining when workloads shift or schemas change. Recent LLM-based query optimizers leverage pre-trained and fine-tuned LLMs to mitigate these challenges. Nevertheless, they neglect LLMs' in-context learning and execution records as feedback for continuous evolution. In this paper, we present SEFRQO, a S elf- E volving F ine-tuned R AG-based Q uery O ptimizer. SEFRQO mitigates the cold-start problem of LQOs by continuously learning from execution feedback via a Retrieval-Augmented Generation (RAG) framework. We employ both supervised fine-tuning and reinforcement fine-tuning to prepare the LLM to produce syntactically correct and performance-efficient query hints. Moreover, SEFRQO leverages the LLM's in-context learning capabilities by dynamically constructing prompts with references to similar queries and the historical execution record of the same query. This self-evolving paradigm iteratively optimizes the prompt to minimize query execution latency. Evaluations show that SEFRQO outperforms state-of-the-art LQOs, achieving up to 65.05% and 93.57% reductions in query latency on the CEB and Stack workloads, respectively, compared to PostgreSQL.

Query optimization is critical in relational databases. Recently, numerous Learned Query Optimizers (LQOs) have been proposed, demonstrating superior performance over traditional hand-crafted query optimizers after short training periods. However, the opacity and instability of machine learning models have limited their practical applications. To address this issue, we are the first to formulate the LQO verification as a Conformal Prediction (CP) problem. We first construct a CP model and obtain user-controlled bounded ranges for the actual latency of LQO plans before execution. Then, we introduce CP-based runtime verification along with violation handling to ensure performance prior to execution. For both scenarios, we further extend our framework to handle distribution shifts in the dynamic environment using adaptive CP approaches. Finally, we present CP-guided plan search, which uses actual latency upper bounds from CP to heuristically guide query plan construction. We integrated our verification framework into three LQOs (Balsa, Lero, and RTOS) and conducted evaluations on several workloads. Experimental results demonstrate that our method is both accurate and efficient. Our CP-based approaches achieve tight upper bounds, reliably detect and handle violations. Adaptive CP maintains accurate confidence levels even in the presence of distribution shifts, and the CP-guided plan search improves both query plan quality (up to 9.84x) and planning time, with a reduction of up to 74.4% for a single query and 9.96% across all test queries from trained LQOs.

The join order (JO) optimization problem is a key challenge in query optimization. Classical approaches can compute the optimal solution for smaller queries. For larger queries, some heuristic methods trade off plan quality to reduce the exponential search space. Recently, quantum-based methods have been proposed to leverage quantum mechanisms to accelerate exploration; however, encoding problem-specific constraints as penalty terms introduces extra overhead. Moreover, quantum-inspired methods on classical hardware do not harness the true advantages of quantum computation. Furthermore, these methods remain at the simulation stage. In this demonstration, we present the first Quantum-augmented Query Optimizer ( Q 2 O ) that integrates a hybrid quantum-classical approach to solve the JO problem in a real database setup. This demonstration allows conference attendees to interact directly with Q 2 O by executing queries and viewing detailed execution results (Scenario 1). Users can easily compare plan quality between Q 2 O and PostgreSQL (Scenario 2). Additionally, they can experiment with quantum parameters to observe their impact (Scenario 3). Experimental results show that the query plan generated by our approach achieves up to a 13x speedup in query execution. The video corresponding to this demonstration is available at this link.

Query optimizers are crucial for the performance of database systems. Recently, many learned query optimizers (LQOs) have demonstrated significant performance improvements over traditional optimizers. However, most of them operate under a limited assumption: a static query environment. This limitation prevents them from effectively handling complex, dynamic query environments in real-world scenarios. Extensive retraining can lead to the well-known catastrophic forgetting problem which reduces the LQO generalizability over time. In this paper, we address this limitation and introduce LIMAO (Lifelong Modular Learned Query Optimizer), a framework for lifelong learning of plan cost prediction that can be seamlessly integrated into existing LQOs. LIMAO leverages a modular lifelong learning technique, an attention-based neural network composition architecture, and an efficient training paradigm designed to retain prior knowledge while continuously adapting to new environments. We implement LIMAO in two LQOs, showing that our approach is agnostic to underlying engines. Experimental results show that LIMAO significantly enhances the performance of LQOs, achieving up to a 40% improvement in query execution time and reducing the variance of execution time by up to 60% under dynamic workloads. By leveraging a precise and self-consistent design, LIMAO effectively mitigates catastrophic forgetting, ensuring stable and reliable plan quality over time. Compared to Postgres, LIMAO achieves up to a 4× speedup on selected benchmarks, highlighting its practical advantages in real-world query optimization.

This paper demonstrates Pyneapple-L, an open-source library designed to enhance scalable spatial analysis through learning-based techniques. Through collaboration with social scientists and domain experts, we identify scalability challenges inherent in conventional spatial analysis methods, particularly as the data size increases. Pyneapple-L addresses these challenges by leveraging learning-based models to offer scalable solutions. We demonstrate two modules: scalable learning of spatial hotspots along spatial networks and augmented geographically weighted regression. To showcase Pyneapple-L, we have developed a user-friendly frontend web application to interact with different datasets, algorithms, model configurations, and visualize outcomes on interactive maps that support both broad and analytical views.

We show how to utilize machine learning approaches to improve sliding window algorithms for approximate frequency estimation problems, under the “algorithms with predictions” framework. In this dynamic environment, previous learning-augmented algorithms are less effective, since properties in sliding window resolution can differ significantly from the properties of the entire stream. Our focus is on the benefits of predicting and filtering out items with large next arrival times – that is, there is a large gap until their next appearance – from the stream, which we show improves the memory-accuracy tradeoffs significantly. We provide theorems that provide insight into how and by how much our technique can improve the sliding window algorithm, as well as experimental results using real-world data sets. Our work demonstrates that predictors can be useful in the challenging sliding window setting.

Video is becoming a major part of contemporary data collection. It is increasingly important to process video selection queries --- selecting videos that contain target objects. Advances in neural networks allow us to detect the objects in an image, and thereby offer query systems to examine the content of the video. Unfortunately, neural network-based approaches have long inference times. Processing this type of query through a standard scan would be time-consuming and would involve applying complex detectors to numerous irrelevant videos. It is tempting to try to improve query times by computing an index in advance. But unfortunately, many frames will never be beneficial for any query. Time spent processing them, whether at index time or at query time, is simply wasted computation. We propose a novel index mechanism to optimize video selection queries with commonsense knowledge. Commonsense knowledge consists of fundamental information about the world, such as the fact that a tennis racket is a tool designed for hitting a tennis ball. To save computation, an inexpensive but lossy index can be intentionally created, but this may result in missed target objects and suboptimal query time performance. Our mechanism addresses this issue by constructing probabilistic models from commonsense knowledge to patch the lossy index and then prioritizing predicate-related videos at query time. This method can achieve significant performance improvements comparable to those of a full index while keeping the construction costs of a lossy index. We describe our prototype system, Paine, plus experiments on two video corpora. We show our best optimization method can process up to 97.79% fewer videos compared to baselines. Even the model constructed without any video content can yield a 75.39% improvement over baselines.

Causal inference can quantify cause-effect relationships in domains as varied as medicine, economics and public policy. Production computer systems exhibit a similar level of complexity and a recurring need to diagnose problems quickly. However, systems are only observed imperfectly, often via long, messy, semi-structured logs. In this work, we want to accelerate large systems debugging by applying causal inference over logs, enabling engineers to diagnose problems and assess interventions in a principled manner. Our framework achieves this through two human-in-the-loop modules: (1) The Candidate Cause Ranker , through which one can determine the causes of a variable without running a full causal discovery algorithm; and (2) the Interactive Causal Graph Refiner , which helps engineers compute an unbiased estimation of their effect of interest without extensive manual causal graph verification. Both modules are powered by the insight that only part of the causal graph of the system is needed to correctly quantify a given effect of interest. We also provide a data preparation pipeline, the Log Converter , which transforms raw, messy, real-world logs into an appropriate tabular input for causal inference, using methods drawn from data transformation, cleaning, and extraction. We evaluate LOGos, a prototype implementation, on both real-world and synthetic logs and find that: (1) The Candidate Cause Ranker achieved an average precision 1.08×-18× higher than the baselines, in interactive time; (2) The Interactive Causal Graph Refiner required a number of causal judgments 1.61 × - 16.83× lower than the baselines; and (3) The latency of Log Converter scaled linearly with three measures of the complexity of a log: length, distinct templates, and fraction of tokens that are variables.

Because video is becoming more popular and constitutes a major part of data collection, we have the need to process video selection queries --- selecting videos that contain target objects. However, a naïve scan of a video corpus without optimization would be extremely inefficient due to applying complex detectors to irrelevant videos. This demo presents Paine; a video query system that employs a novel index mechanism to optimize video selection queries via commonsense knowledge. Paine samples video frames to build an inexpensive lossy index, then leverages probabilistic models based on existing commonsense knowledge sources to capture the semantic-level correlation among video frames, thereby allowing Paine to predict the content of unindexed video. These models can predict which videos are likely to satisfy selection predicates so as to avoid Paine from processing irrelevant videos. We will demonstrate a system prototype of Paine for accelerating the processing of video selection queries, allowing VLDB'23 participants to use the Paine interface to run queries. Users can compare Paine with the baseline, the SCAN method.

In-memory join is an essential operator in any database engine. It has been extensively investigated in the database literature. In this paper, we study whether exploiting the CDF-based learned models to boost the join performance is practical. To the best of our knowledge, we are the first to fill this gap. We investigate the usage of CDF-based models and learned indexes (e.g., Recursive Model Index (RMI) and RadixSpline) in the three join categories; indexed nested loop join (INLJ), sort-based joins (SJ) and hash-based joins (HJ). Our study shows that there is room to improve the performance of the three join categories through our proposed optimized learned variants. Our experimental analysis showed that these optimized learned variants outperform the state-of-the-art techniques in many scenarios and with different datasets.

Query scheduling is a crucial task for analytical database systems that can greatly affect the query latency. However, existing scheduling approaches are based on heuristics and not optimal. A recent trial proposed to use reinforcement learning for automatically learning end-to-end scheduling policies. However, such trial was not capable of considering the database-specific characteristics (e.g., operator types, pipelining), and hence becomes not efficient for analytical database systems. In this paper, we try to fill this gap and introduce LSched (Learned Scheduler), a fully learned workload-aware query scheduler for in-memory analytical database systems. LSched provides an efficient inter-query and intra-query scheduling for dynamic analytical workloads (i.e., different queries can arrive/depart at any time). We integrated LSched with an efficient in-memory analytical database system, and evaluated it with TPCH, SSB, and JOB benchmarks. Our evaluation shows that LSched improves over the performance of existing state-of-the-art query schedulers and heuristic-based ones by at least 35% and 50% in both streaming and batching query workloads.

Hashing is a fundamental operation in database management, playing a key role in the implementation of numerous core database data structures and algorithms. Traditional hash functions aim to mimic a function that maps a key to a random value, which can result in collisions, where multiple keys are mapped to the same value. There are many well-known schemes like chaining, probing, and cuckoo hashing to handle collisions. In this work, we aim to study if using learned models instead of traditional hash functions can reduce collisions and whether such a reduction translates to improved performance, particularly for indexing and joins. We show that learned models reduce collisions in some cases, which depend on how the data is distributed. To evaluate the effectiveness of learned models as hash function, we test them with bucket chaining, linear probing, and cuckoo hash tables. We find that learned models can (1) yield a 1.4x lower probe latency, and (2) reduce the non-partitioned hash join runtime with 28% over the next best baseline for certain datasets. On the other hand, if the data distribution is not suitable, we either do not see gains or see worse performance. In summary, we find that learned models can indeed outperform hash functions, but only for certain data distributions.

The proliferation in amounts of generated data has propelled the rise of scalable machine learning solutions to efficiently analyze and extract useful insights from such data. Meanwhile, spatial data has become ubiquitous, e.g., GPS data, with increasingly sheer sizes in recent years. The applications of big spatial data span a wide spectrum of interests including tracking infectious disease, climate change simulation, drug addiction, among others. Consequently, major research efforts are exerted to support efficient analysis and intelligence inside these applications by either providing spatial extensions to existing machine learning solutions or building new solutions from scratch. In this 90-minutes seminar, we comprehensively review the state-of-the-art work in the intersection of machine learning and big spatial data. We cover existing research efforts and challenges in three major areas of machine learning, namely, data analysis, deep learning and statistical inference. We also discuss the existing end-to-end systems, and highlight open problems and challenges for future research in this area.

This paper presents Sya; the first spatial probabilistic knowledge base construction system, based on Markov Logic Networks (MLN). Sya injects the awareness of spatial relationships inside the MLN grounding and inference phases, which are the pillars of the knowledge base construction process, and hence results in a better knowledge base output. In particular, Sya generates a probabilistic model that captures both logical and spatial correlations among knowledge base relations. Sya provides a simple spatial high-level language, a spatial variation of factor graph, a spatial rules-query translator, and a spatially-equipped statistical inference technique to infer the factual scores of relations. In addition, Sya provides an optimization that ensures scalable grounding and inference for large-scale knowledge bases. Experimental evidence, based on building two real knowledge bases with spatial nature, shows that Sya can achieve 70% higher F1-score on average over the state-of-the-art DeepDive system, while achieving at least 20% reduction in the execution times.

The proliferation in amounts of generated data has propelled the rise of scalable machine learning solutions to efficiently analyze and extract useful insights from such data. Meanwhile, spatial data has become ubiquitous, e.g., GPS data, with increasingly sheer sizes in recent years. The applications of big spatial data span a wide spectrum of interests including tracking infectious disease, climate change simulation, drug addiction, among others. Consequently, major research efforts are exerted to support efficient analysis and intelligence inside these applications by either providing spatial extensions to existing machine learning solutions or building new solutions from scratch. In this 90-minutes tutorial, we comprehensively review the state-of-the-art work in the intersection of machine learning and big spatial data. We cover existing research efforts and challenges in three major areas of machine learning, namely, data analysis, deep learning and statistical inference. We also discuss the existing end-to-end systems, and highlight open problems and challenges for future research in this area.

No abstract available.

Today, a modern data center hosts a wide variety of applications comprising batch, interactive, machine learning, and streaming applications. In this paper, we factor out the commonalities in a large majority of these applications, into a generic dataflow layer called Common Runtime for Applications (CRA). In parallel, another trend, with containerization technologies (e.g., Docker), has taken a serious hold on cloud-scale data centers, with direct implications on building next generation of data center applications. Container orchestrators (e.g., Kubernetes) have made deployment a lot easy, and they solve many infrastructure level problems, e.g., service discovery, auto-restart, and replication. For best in class performance, there is a need to marry the next generation applications with containerization technologies. To that end, CRA leverages and builds upon the containerization and resource orchestration capabilities of Kubernetes/Docker, and makes it easy to build a wide range of cloud-edge applications on top. To the best of our knowledge, we are the first to present a cloud native runtime for building data center applications. We show the efficiency of CRA through various micro-benchmarking experiments.

The proliferation in amounts of generated data has propelled the rise of scalable machine learning solutions to efficiently analyze and extract useful insights from such data. Meanwhile, spatial data has become ubiquitous, e.g., GPS data, with increasingly sheer sizes in recent years. The applications of big spatial data span a wide spectrum of interests including tracking infectious disease, climate change simulation, drug addiction, among others. Consequently, major research efforts are exerted to support efficient analysis and intelligence inside these applications by either providing spatial extensions to existing machine learning solutions or building new solutions from scratch. In this 90-minutes tutorial, we comprehensively review the state-of-the-art work in the intersection of machine learning and big spatial data. We cover existing research efforts and challenges in three major areas of machine learning, namely, data analysis, deep learning and statistical inference, as well as two advanced spatial machine learning tasks, namely, spatial features extraction and spatial sampling. We also highlight open problems and challenges for future research in this area.

The current explosion in spatial data raises the need for efficient spatial analysis tools to extract useful information from such data. However, existing tools are neither generic nor scalable when dealing with big spatial data. This demo presents Flash ; a framework for generic and scalable spatial data analysis, with a special focus on spatial probabilistic graphical modelling (SPGM). Flash exploits Markov Logic Networks (MLN) to express SPGM as a set of declarative logical rules. In addition, it provides spatial variations of the scalable RDBMS-based learning and inference techniques of MLN to efficiently perform SPGM predictions. To show Flash effectiveness, we demonstrate three applications that use Flash in their SPGM: (1) Bird monitoring, (2) Safety analysis, and (3) Land use change tracking.

Predicting the presence or absence of spatial phenomena has been of great interest to scientists pursuing research in several applications including epidemic diseases detection, species occurrence prediction and earth observation. In this operation, a geographical space is divided by a two-dimensional grid, where the prediction (i.e, either 0 or 1) is performed at each cell in the grid. A common approach to solve this problem is to build spatial logistic regression models (a.k.a autologistic models) that estimate the prediction at any location based on a set of predictors (i.e., features) at this location and predictions from neighboring locations. Unfortunately, existing methods to build autologistic models are computationally expensive and do not scale up for large-scale grid data (e.g., fine-grained satellite images). This paper introduces TurboReg, a scalable framework to build autologistic models for predicting large-scale spatial phenomena. TurboReg considers both the accuracy and efficiency aspects when learning the regression model parameters. TurboReg is built on top of Markov Logic Network (MLN), a scalable statistical learning framework, where its internals and data structures are optimized to process spatial data. A set of experiments using large real and synthetic data show that TurboReg achieves at least three orders of magnitude performance gain over existing methods while preserving the model accuracy.

This demo presents Sya; the first full-fledged spatial probabilistic knowledge base construction system. Sya is a comprehensive extension to the DeepDive system that enables exploiting the spatial relationships between extracted relations during the knowledge base construction process, and hence results in a better knowledge base output. Sya runs existing DeepDive programs as is, yet, it extracts more accurate relations than DeepDive when dealing with input data that have spatial attributes. Sya employs a simple spatial high-level language, a rule-based spatial SQL query engine, a spatially-indexed probabilistic graphical model, and an adapted spatial statistical inference technique to infer the factual scores of relations. We demonstrate a real system prototype of Sya, showing a case study of constructing a crime knowledge base. The demonstration shows to the audience the internal steps of building the knowledge base, as well as a comparison with the output of DeepDive.

This paper provides the first attempt for a full-fledged query optimizer for MapReduce-based spatial join algorithms. The optimizer develops its own taxonomy that covers almost all possible ways of doing a spatial join for any two input datasets. The optimizer comes in two flavors; cost-based and rule-based. Given two input data sets, the cost-based query optimizer evaluates the costs of all possible options in the developed taxonomy, and selects the one with the lowest cost. The rule-based query optimizer abstracts the developed cost models of the cost-based optimizer into a set of simple easy-to-check heuristic rules. Then, it applies its rules to select the lowest cost option. Both query optimizers are deployed and experimentally evaluated inside a widely used open-source MapReduce-based big spatial data system. Exhaustive experiments show that both query optimizers are always successful in taking the right decision for spatially joining any two datasets of up to 500GB each.

Device-free (DF) localization in WLANs has been introduced as a value-added service that allows tracking of indoor entities that do not carry any devices. Previous work in DF WLAN localization focused on the tracking of a single entity due to the intractability of the multi-entity tracking problem whose complexity grows exponentially with the number of humans being tracked. In this paper, we introduce ACE: a system that uses a probabilistic energy-minimization framework that combines a conditional random field with a Markov model to capture the temporal and spatial relations between the entities' poses. A novel cross-calibration technique is introduced to reduce the calibration overhead of multiple entities to linear, regardless of the number of humans being tracked. We design an efficient energy-minimization function that can be mapped to a binary graph-cut problem whose solution has a linear complexity on average and a third order polynomial in the worst case. We further employ clustering on the estimated location candidates to reduce outliers and obtain more accurate tracking in the continuous space. Experimental evaluation in two typical testbeds, with a side-by-side comparison with the state-of-the-art, shows that ACE can achieve a multi-entity tracking accuracy of less than 1.3 m. This corresponds to at least 11.8 percent, and up to 33 percent, enhancement in median distance error over the state-of-the-art DF localization systems. In addition, ACEcan estimate the number of entities correctly to within one difference error for 100 percent of the time. This highlights that ACE achieves its goals of having an accurate and efficient multi-entity indoors localization.

Device-free (DF) indoor localization has grasped great attention recently as a value-added service to the already installed WiFi infrastructure as it allows the tracking of entities that do not carry any devices nor participate actively in the localization process. Current approaches, however, require a relatively large number of wireless streams, i.e. transmitter-receiver pairs, which is not available in many typical scenarios, such as home monitoring. In this paper, we introduce MonoPHY as an accurate mono-stream device-free WLAN localization system. MonoPHY leverages the physical layer information of WiFi networks supported by the IEEE 802.11n standard to provide accurate DF localization with only one stream. In particular, MonoPHY leverages both the low-level Channel State Information and the MIMO information to capture the human effect on signal strength. Experimental evaluation in a typical apartment, with a side-by-side comparison with the state-of-the-art, shows that MonoPHY can achieve an accuracy of 1.36m. This corresponds to at least 48% enhancement in median distance error over the state-of-the-art DF localization systems using a single stream only.

We introduce Spot as an accurate and efficient system for multi-entity device-free (DF) detection and tracking. Current state-of-the-art systems focused on the tracking of a single entity because of the intractability of the multi-entity case that leads to exponential complexity. Spot provides a novel cross-calibration technique that reduces the overhead of multiple entities calibration from exponential to linear. Spot also captures the spatial relations between the entities' poses into a probabilistic energy minimization framework via a conditional random field model. The designed energy minimization function is solved by a binary graph-cut algorithm. We evaluate our system using a typical testbed and show that Spot can achieve a multi-entity median tracking error of less than 1.44m. This corresponds to 108.33% enhancement in median distance error over the state-of-the-art DF localization systems, which can only track a single entity. In addition, Spot can estimate the number of entities correctly to within one difference error with 92% accuracy. This highlights that Spot achieves its goals of having an accurate and efficient software-only DF tracking solution of multiple entities in indoor environments.

Nowadays, satisfying user needs has become the main challenge in a variety of web applications. Recommender systems play a major role in that direction. However, as most of the information is present in a textual form, recommender systems face the challenge of efficiently analyzing huge amounts of text. The usage of semantic-based analysis has gained much interest in recent years. The emergence of ontologies has yet facilitated semantic interpretation of text. However, relying on an ontology for performing the semantic analysis requires too much effort to construct and maintain the used ontologies. Besides, the currently known ontologies cover a small number of the world's concepts especially when a non-domain-specific concepts are needed. This paper proposes the use of Wikipedia as ontology to solve the problems of using traditional ontologies for the text analysis in text-based recommendation systems. A full system model that unifies semantic-based analysis with a collaborative via content recommendation system is presented.