arXiv subjects

Nada Ali

Publications and source records attributed to Nada Ali.

7 recordsLinked to original sources

arXiv2026

Data-Driven Telecom Marketing Optimization: A Machine Learning-Based Churn Prediction and Customer Segmentation Framework

Customer churn is a major challenge for telecommunication companies, directly eroding revenue and long term customer relationships. Traditional retention programs rely on generic, not personalized incentives and lack the precision to identify high risk customers before they leave. This paper presents a data driven marketing optimization framework integrating machine learning based churn prediction, customer segmentation combining churn risk with customer value, and tailored, segment specific marketing and Return on Investment ROI strategies. Using the IBM Telco Customer Churn dataset with 7043 customers and 21 features, three gradient boosting ensembles, XGBoost, LightGBM, and CatBoost, were trained and tuned via randomized search with stratified 5 fold cross validation, class weighting, and F1 score driven decision threshold optimization to counter a class imbalance of 73.4% versus 26.6%. CatBoost was selected as the deployment model, achieving 77.68% accuracy, an F1 score of 0.6366, a PR AUC of 0.6553, and a ROC AUC of 0.8403 on the held out test set. Customers were partitioned with K Means clustering, validated via the Elbow method and visualized with Principal Component Analysis, into High, Medium, and Low Value segments, cross tabulated against churn risk labels to define four actionable clusters. Segment specific retention, upsell, and engagement strategies were designed for each cluster, and a theoretical ROI and CLV framework quantifies the financial impact of the proposed interventions. The pipeline was operationalized in an interactive Streamlit web application allowing marketing teams to upload data, filter by segment, visualize churn drivers via SHAP, and download automated segment reports. Results confirm that combining predictive churn modeling with value aware segmentation yields more actionable and profitable marketing decisions than churn prediction alone.

cs.LG↗

arXiv2026

Detection of patterns in a discrete-outcome sensor network

A discrete outcome quantum sensor network is one in which we are only interested in which detectors are activated. This can be studied in either the strong or weak interaction regime. If the detectors interact strongly with the environment, it is possible to definitely find which ones were activated. If the interaction is weaker, there is a possibility of making an error, and the object is to minimize the probability of this happening. Here we will be interested in this weaker interaction regime. We will also assume that only certain patterns of detectors will be activated, different patterns being translated versions of a fundamental one. Our object will be to find which pattern has been activated. We will look at both one and two-dimensional detector arrays and make use of techniques from minimum-error state discrimination.

quant-ph↗

arXiv2025

Slope gap distribution of the double heptagon and an algorithm for determining winning vectors

In this paper, we study the distribution of renormalized gaps between slopes of saddle connections on translation surfaces. Specifically, we describe a procedure for finding the "winning holonomy vectors" as defined by Kumanduri-Sanchez-Wang in arXiv:2102.10069, which constitutes a key step in calculating the slope gap distribution for an arbitrary Veech surface. We then apply this method to explicitly compute the gap distribution for the regular double heptagon translation surface. This extends work of Athreya-Chaika-Lelievre in arXiv:1308.4203 on the gap distribution for the "golden L" translation surface, which is equivalent to the regular double pentagon surface.

math.DS↗

arXiv2025

Interference and Measurement: Changing amplitude phase information to amplitude magnitude information

There are quantum procedures that encode the solutions to a problem in the phases of quantum amplitudes. This happens in some quantum optimization algorithms in which the value of a function to be maximized or minimized is represented by a phase. An example of this is the QAOA algorithm for the MaxCut problem in which one encodes the number of edges connecting the sets resulting from a partition of the vertices of a graph into phases of amplitudes of a quantum state. Another is the minimum vertex cover problem in which the number of edges included in the cover is encoded in phases. Here we want to see what can be done if we only use simple aspects of quantum mechanics, interference and measurement, to manipulate the magnitudes of the amplitudes whose phases encode the relevant information. The idea is to use constructive interference to enhance the amplitudes that contain useful information and destructive interference to suppress those that do not. We examine examples, both analytically and numerically. We also show how the results of sequences of measurements can be used to gain information about the landscape of solutions.

quant-ph↗

arXiv2024

State learning from pairs of states

Suppose you receive a sequence of qubits where each qubit is guaranteed to be in one of two pure states, but you do not know what those states are. Your task is to determine the states. This can be viewed as a kind of quantum state learning -- or quantum state estimation. A problem is that, without more information, all that can be determined is the density matrix of the sequence and, in general, density matrices can be decomposed into pure states in many different ways. To solve the problem, additional information, either classical or quantum, is required. We show that if an additional copy of each qubit is supplied -- that is, one receives pairs of qubits, both in the same state, rather than single qubits -- the task can be accomplished. This is possible because the mixed two-qubit state has a unique decomposition into pure product states. For illustration, we simulate numerically the symmetric, informationally complete measurement of a sequence of qubit pairs and show that the unknown states and their respective probabilities of occurrence can be inferred from the data with high accuracy. Finally, we propose an experiment that employs a product measurement and can be realized with existing technology, and we demonstrate how the data tell us the states and their probabilities. We find that it is enough to detect a few thousand qubit pairs.

quant-ph↗

arXiv2024

Discrete-outcome sensor networks: Multiple detection events and grouping detectors

Quantum sensor networks have often been studied in order to determine how accurately they can determine a parameter, such as the strength of a magnetic field, at one of the detectors. A more coarse-grained approach is to try to simply determine whether a detector has interacted with a signal or not, and which detector it was. Such discrete-outcome quantum sensor networks, discrete in the sense that we are seeking answers to yes-no questions, are what we study here. One issue is what is a good initial state for the network, and, in particular, should it be entangled or not. Earlier we looked at the case when only one detector interacted, and here we extend that study in two ways. First, we allow more that one detector to interact, and second, we examine the effect of grouping the detectors. When the detectors are grouped we are only interested in which group contained interacting detectors and not in which individual detectors within a group interacted. We find that in the case of grouping detectors, entangled initial states can be helpful.

quant-ph↗

arXiv2023

Two entanglement conditions and their connection to negativity

We examine two conditions that can be used to detect bipartite entanglement, and show that they can be used to provide lower bounds on the negativity of states. We begin with two-qubit states, and then show how what was done there can be extended to more general states. The resulting bounds are then studied by means of a number of examples. We also show that if one has some knowledge of the Schmidt vectors of a state, better bounds can be found.

quant-ph↗