Trending November 2023 # Issues Of Validity In Evolutionary Psychology Research # Suggested December 2023 # Top 14 Popular

You are reading the article Issues Of Validity In Evolutionary Psychology Research updated in November 2023 on the website We hope that the information we have shared is helpful to you. If you find the content interesting and meaningful, please share it with your friends and continue to follow and support us for the latest updates. Suggested December 2023 Issues Of Validity In Evolutionary Psychology Research

Evolutionary psychology is a relatively new field that seeks to understand human behavior in terms of how it evolved. Evolutionary psychology assumes that the human mind has evolved through natural selection to solve specific adaptive problems faced by our ancestors in the environment in which they lived. The degree to which a test or assessment reliably measures what it promises to measure is called its validity. Validity issues can develop when a test or measurement fails to capture the desired concept accurately.

Issues of Validity in Evolutionary Psychology Research

In evolutionary psychology, validity refers to the extent to which a hypothesis or theory is supported by evidence from evolutionary biology and the principles of natural selection. Although it has generated much interest and excitement, there are some concerns about the validity of evolutionary psychology research. In evolutionary psychology, researchers frequently use comparative research methods, such as cross-cultural studies and studies of non-human animals, to test whether a particular trait or behavior is shared across different species and cultures or is unique to humans.

Furthermore, evolutionary psychology researchers frequently employ evolutionary models and mathematical simulations to assess the plausibility of their ideas and forecast how specific characteristics or behaviors may have developed.

Brief Issues in Evolutionary Psychology Researches

While evolutionary psychology has influenced our understanding of human behavior and cognition, concerns and criticisms exist regarding its validity. One issue is that evolutionary psychology ideas are difficult to explicitly test since they frequently require making inferences about events that occurred in the distant past. This might lead to disagreements about the validity of various evolutionary explanations for a specific behavior or trait.

Another concern is that evolutionary psychology frequently depends on assumptions about the universality of human behavior and cognition without considering cultural and social influences that may affect these phenomena. As a result, substantial individual and cultural differences in behavior and cognition may be overlooked. These are characterized as Sample Bias, Environment Role, Over-Generalization, Self-Reporting Bias, and Lack of Experimental Evidence.

Sample Bias

The convenience samples of individuals in evolutionary psychology studies can limit the generalizability of outcomes. The emphasis on studying WEIRD (Western, Educated, Industrialised, Rich, and Democratic) groups and the over-representation of college students and young people can result in findings that do not apply to other populations. For example, suppose a study focuses on college-age women in the United States.

In that case, the conclusions may be inapplicable to other populations, such as men, older adults, or persons from different cultural backgrounds. Different researchers have discovered that the findings of evolutionary psychology investigations are frequently not duplicated in different nations, implying a lack of generalizability. For example, a study on the association between physical appearance and partner selection discovered that the results varied between countries. Physical appearance was a more relevant component in mate selection in some countries than others. This implies that the study’s findings may not apply to all populations.

Environment Role

Evolutionary psychologists frequently emphasize the significance of biology in behavior without considering the influence of environmental circumstances. This limited perspective might lead to the oversimplification of complicated behaviors. For example, in a study of mate selection, evolutionary psychologists may focus on biological attributes such as physical appearance while ignoring other contextual factors such as education or income that may be essential. This limited perspective might lead to the oversimplification of complicated behaviors. Also, a study found that men are more attracted to physical attractiveness in a mate than women, even when other contextual characteristics such as education level or income are not considered. These criteria may be more essential in deciding mate selection than physical appearance.


Based on a small number of studies and participants, evolutionary psychologists can occasionally make sweeping generalizations about the behavior of all people. This can result in results that do not reflect the diversity of human behavior. A study, for example, may conclude that males are more attracted to physical attractiveness in a mate than women without considering individual differences or cultural context.

This can result in results that do not reflect the diversity of human behavior. In another instance, researchers conclude that males are more attracted to physical attractiveness in a spouse than women, ignoring cultural or individual preferences. In actuality, mate selection preferences vary greatly depending on cultural context and individual preferences.

Self-Reporting Bias

Much research in evolutionary psychology relies on self-report data, which can be erroneous owing to social desirability bias or other issues. People may be more prone to describe socially desirable behaviors like compassion or empathy than they are to engage in the behaviors themselves.

This can result in the overestimation or underestimation of certain behaviors. A study, for example, may discover that people report higher levels of generosity than they exhibit in a lab context. This shows that, in some circumstances, self-report data may not accurately assess behavior.

Lack of Experimental Evidence

Many kinds of research in evolutionary psychology are correlational, limiting the capacity to make firm causal conclusions. For example, a study finding a correlation between physical attractiveness and partner selection does not establish that one causes the other. More experimental evidence is required before drawing firm conclusions regarding the influence of evolution on behavior. For example, a study may indicate that men choose more physically appealing mates, but this does not prove that physical attractiveness is the reason for mate selection. An experiment that modified physical beauty and analyzed its effect on mate selection would be required to establish more specific results.


These concerns emphasize the importance of evolutionary psychology researchers employing rigorous empirical methodologies while acknowledging their theories’ limitations. While evolutionary psychology has created many intriguing theories about human behavior, these hypotheses must be empirically validated and rigorously scrutinized to ensure their validity and reliability.

You're reading Issues Of Validity In Evolutionary Psychology Research

Ethics In Psychology: Meaning And Application

APA outlines general ethical principles that ensure the safety and well-being of the participants involved in a study. In addition, APA has a list of ethical standards, including protection against harm, harassment, and discrimination.

What are Ethics?

Ethics are defined as moral or philosophical codes that focus on the concept of the things that are right and which are wrong. An ethical approach guides individuals’ behavior toward what is right and wrong. In the field of Psychology, ethical guidelines are adopted to ensure that the patients who are given therapy and the participants involved in research conducted by Psychology researchers do not face any negative consequences as a result of their participation. Ethics in Psychology are of utmost importance and protect the right and dignity of individuals. Ethics in Psychology apply to various fields, including therapy, research, education, and publication.

Five general principles of APA

Following are the five general principles of APA −

Beneficence and Non-maleficence − The first principle of APA states that during the study or research, psychologists and researchers need to safeguard the rights and welfare of the individuals involved in the study and maximize their benefits. This principle states that researchers must work freely from biases and prejudices. Hence, researchers must conduct research without biases that may negatively affect the study.

Fidelity and Responsibility − This principle highlights the importance of conscientiousness in the research and study conducted by psychologists. It is a universally understood principle and is concerned with compliance with ethical principles with colleagues and others in the work network. Ethical misconduct by researchers needs to be pointed out whenever spotted by others, but the respect and dignity of the researcher should not be violated in the process.

Justice − The fourth principle states that fairness and justice should be practiced by all researchers, such that equality is practiced throughout the study and all the participants benefit from the services and the research conducted by the practitioners.

Respect for Rights and Dignity − The fifth principle of APA emphasizes obtaining individuals’ consent before the study’s conduct and safeguarding the autonomy and confidentiality of the participants involved in the study. Psychologists need to be aware of and respect cultural, linguistic, and gender differences and not violate the rights and dignity of the individuals while the research is being conducted.

Ethical considerations in Psychology

The code of ethics given by the American Psychological Association guides the appropriate conduct of psychologists and practitioners. Psychologists often need a specific framework of guidelines that help maintain their ethical practices in their specific professional context. The APA code of ethics clarifies the appropriate professional behavior in the various aspects of practice. Psychologists deal with sensitive situations, and ethical concerns play a vital role in this case.

Informed Consent

Psychologists and practitioners play multiple roles as researchers, therapists, consultants, and educators. For example, while dealing with patients, therapists must inform them about the services offered and what to expect from them. While conducting research, it is important to let participants know about the purpose of the study and the potential risks involved.

Client Welfare

Psychologists must ensure that the services they provide work toward their client’s welfare and do not harm them.


Psychologists must ensure that patients’ sensitive information and details are not shared with a third party and that privacy and confidentiality are not breached.


Therapists should only lie about their areas of expertise and provide services in the areas they are competent in if it is an emergency.

Ethical considerations followed by researchers

Avoiding harm to participants caused during the study.

Conduct the study truthfully and honestly to convey the findings to others.

Include the strengths and weaknesses of the study in the research article.

Collect facts and details accurately before actually carrying out the assessment.

Assure that the participants are provided with informed consent before the conduction of the study and are made aware of their rights.

Maintaining the confidentiality and anonymity of respondents and the results of their study.

Ethical considerations followed by Practicing Psychologists

These are −

Fidelity − Fidelity, or being trustworthy, involves the principles of loyalty, faithfulness, and fulfilling commitments. Practitioners who comply with the principle of fidelity act under the trust and faith invested in them by the client, aim to meet clients’ expectations and honor their clients’ confidentiality and privacy along with safeguarding their dignity.

Autonomy − This principle allows clients the freedom of action and choice. This involves the ability of Psychologists and practitioners to encourage the clients to be self-dependent and make their own decisions wherever possible. The practitioner has the responsibility to help the clients understand the impact of their actions on themselves and society and help them make rational and responsible decisions in the future.

Beneficence − This principle highlights the importance of acting in a way that serves the participants’ best interests and looks after clients’ welfare. Furthermore, it focuses on strictly working within the competence and providing adequate services by trained professionals.

Non-Maleficence − Non-maleficence is the principle that focuses on not causing harm to others at any cost. It focuses on not causing intentional or unintentional harm to others in any way and avoiding social, sexual, and financial exploitation of the clients. The practitioner is responsible for tackling any harm that may befall their clients.

Justice − This principle highlights the importance of treating all clients equally and fairly, irrespective of their cultural, religious, linguistic, and gender differences. Justice does not necessarily imply treating all participants equally but proportionately, and there needs to be a rational explanation for the choice to treat a client differently from others.


It is crucial for Psychology to articulate certain principles of ethics, being a scientific discipline. It gives credibility and respect to the researchers and practicing Psychologists and help in resolving ethical issues which may be ambiguous by proving guidelines and an ethical code of conduct. There are arguments that, in the past, ethical considerations may have been broken, taking an example of Little Albert’s experiment conducted by Watson because it caused harm to the participant of the study. Others argue that unethical researches are a thing of the past Psychology is evolving, and ethics have become much more important in the present than in the past.

League Of Legends Patch 10.1 Reported Issues

League of Legends Patch 10.1 Reported Issues




True gamers use the best gaming browser: Opera GX

Opera GX is a special version of the famous Opera browser that is built specifically to fulfill gamer’s needs. Packed with unique features, Opera GX will help you get the most out of gaming and browsing everyday:

CPU, RAM and Network limiter with hot tab killer

Integrated with Twitch, Discord, Instagram, Twitter and Messengers directly

Built-in sound controls and custom music

Custom color themes by Razer Chroma and force dark pages

Free VPN and Ad blocker

Download Opera GX

Patch 10.1 of the critically acclaimed MOBA League of Legends has hit live, kick-starting the 2023 season of this E-sports giant.

While it does bring plenty of changes, it comes with its own fair share of issues. Since the patch is quite a large one, it goes without saying that a lot of new content has been added as well, creating an ideal scenario for bugs to appear.

That is why we’ve compiled a list of what we think are the most common and serious bugs currently found in League of Legends patch 10.1.

Common bugs and issues in League of Legends patch 10.1 1. Users can’t see their friend list

It would seem that the game’s UI is bugged, and users cannot see any of the information that would usually be found on the right side of the menu.

It would seem that in order for users to encounter this bug, they would have to swap between EUW to EUNE servers during the patching process.

Unfortunately, repairing the client doesn’t work.

Thus, it would be a good idea to wait until the patch is complete before attempting to change the server. Additionally, it seems that if you just play a match, the client will go back to normal on itself.

2. Players cannot pick their Rune pages

Apparently, many players cannot access their created rune pages, and their default rune pages were changed as well.

Unfortunately, no one knows what triggers this bug, nor if there is any known solution to it.

3. Players receive a “password required” message after matches

Apparently, some users receive a message that states the following

A password is required to enter this room” dialogue box appears after every game.

Normally, this would be the message users would receive when trying to join a custom match to which the owner has assigned a password.

4. Users are experiencing packet losses

In general terms, packet losses occur when one or more packets of data traveling across a computer network fail to reach their destination.

In League of Legends, packet losses may lead to NPCs and enemy champions appearing to phase in and out from on destination, as if teleporting.

Unfortunately, there is no current fix for this issue. However, check that your Internet connection is working appropriately may fix this.

5. In-game chat is bugged

Users have reported that in-game, other player’s names appear as random numbers and letters. This makes it extremely difficult to communicate in-game.

Unfortunately, no one knows what causes this, or how to fix it.


That about wraps up the list of the most common League of Legends Patch 10.1 bugs that players have found so far.

While a lot of them are indeed game-breakers, Riot always manages to deliver on their end when it comes to fixing them.

As such, players should expect to see these issues fixed by the next patch update.


Still experiencing issues?

Was this page helpful?


Start a conversation

Discussion With Dr. Prabhakar Raghavan, Head Of Research At Yahoo

The Hindu Business Line has an interesting discussion on search, user intent, micro-economics of search and user generated meta data with Dr. Prabhakar Raghavan, Head of Research at Yahoo.

If you’ve ever had the pleasure of hearing Dr. Raghavan speak, you’d know how the man’s enthralling passion for search and his research team invigorates the conversation with drive and confidence.

If you haven’t, this discussion with K. Bharat Kumar of Hindu Business Line is the next best thing.

On Searcher Intent

At Yahoo! type ‘Papa John’s’. It’s a pizza chain in the US. It’s also a publicly traded company. The top result is the home page of the company. That is to be expected. Along with this, I give two more links. I see that it could be an investor interested in the stock. So I give a link to Yahoo! Financials for related information about the company. On the other hand, you could be hungry. So, I give you phone numbers and addresses of outlets nearest your home. I am now trying to resolve possible goals.

So, we sequence your query, divine your intent and get you there as quickly as possible. It serves you and hence us well. So you think we are prescient and you come back again and again.

On Google vs. Yahoo

Google talks about universal search. Microsoft and we are thinking about it. We do not want to give users 10 matching documents but actually craft an experience that fulfils a need.

Airports have three-letter codes. For example, MAA stands for Chennai while SFO stands for San Francisco. If you type these two along with a date, Google gives you matching documents that could be nonsense. At the top, it also fills out a travel form with these two cities with a couple of dates with links to Expedia and the like. Anyone typing that in is almost certain to buy air tickets. So, why not give it to them, instead of some documents?

On User Reviews and Monetary Incentives

It’s not necessarily money as incentive. I should feel good that something useful emerges out of it. In Yahoo! Answers, you get points so you get a rush. Question is how sustainable is that and how much value is generated from the content that is unique and defensible. It’s glib to say that Yahoo! Answers succeeded and that Google shut down its version. Now comes the ’then what?’ question.

The challenge for us is to figure out underlying motives and why people do what they do. Maybe in search there is intent to get task done. Communication is more casual – human nature is to communicate. As to content creation and consumption, human nature plays a role there too. But you need to get the expression to create greater good.

Topic Modeling: Predicting Multiple Tags Of Research Articles Using Onevsrest Strategy

This article was published as a part of the Data Science Blogathon

Recently I participated in an NLP hackathon — “Topic Modeling for Research Articles 2.0”. This hackathon was hosted by the Analytics Vidhya platform as a part of their HackLive initiative. The participants were guided by experts in a 2-hour live session and later on were given a week to compete and climb the leaderboard.

Problem Statement

Given the abstracts for a set of research articles, the task is to predict the tags for each article included in the test set.

The research article abstracts are sourced from the following 4 topics — Computer Science, Mathematics, Physics, Statistics. Each article can possibly have multiple tags among 25 tags like Number Theory, Applications, Artificial Intelligence, Astrophysics of Galaxies, Information Theory, Materials Science, Machine Learning et al. Submissions are evaluated on micro F1 Score between the predicted and observed tags for each article in the test set.

Complete Problem Statement and the dataset is available here.

Without further ado let’s get started with the code.

Loading and Exploring data

Importing necessary libraries —

%matplotlib inline import numpy as np import pandas as pd import matplotlib.pyplot as plt import seaborn as sns from nltk.tokenize import word_tokenize from chúng tôi import PorterStemmer from sklearn.feature_extraction.text import CountVectorizer from sklearn.feature_extraction.text import TfidfVectorizer from sklearn import metrics from sklearn.metrics import accuracy_score from sklearn.metrics import f1_score

Load train and test data from .csv files into Pandas DataFrame —

train_data = pd.read_csv(‘Train.csv’) test_data = pd.read_csv(‘Test.csv’)

Train and Test data shape —

print(“Train size:”, train_data.shape) print(“Test size:”, test_data.shape)


There are ~ 14k datapoints in the Train dataset and ~6k datapoints in the Test set. Overview of train and test datasets —



As we can see from the train data info, there are 31 columns — 1 column for id, 1 column for Abstract text, 4 columns for topics, these all form our feature variables, and the next 25 columns are class-labels that we have to ‘learn’ for the prediction task.

topic_cols = [‘Computer Science’, ‘Mathematics’, ‘Physics’, ‘Statistics’] target_cols = [‘Analysis of PDEs’, ‘Applications’, ‘Artificial Intelligence’, ‘Astrophysics of Galaxies’, ‘Computation and Language’, ‘Computer Vision and Pattern Recognition’, ‘Cosmology and Nongalactic Astrophysics’, ‘Data Structures and Algorithms’, ‘Differential Geometry’, ‘Earth and Planetary Astrophysics’, ‘Fluid Dynamics’, ‘Information Theory’, ‘Instrumentation and Methods for Astrophysics’, ‘Machine Learning’, ‘Materials Science’, ‘Methodology’, ‘Number Theory’, ‘Optimization and Control’, ‘Representation Theory’, ‘Robotics’, ‘Social and Information Networks’, ‘Statistics Theory’, ‘Strongly Correlated Electrons’, ‘Superconductivity’, ‘Systems and Control’]

How many datapoints have more than 1 tags?

my_list = [] for i in range(train_data.shape[0]): my_list.append(sum(train_data.iloc[i, 6:])) pd.Series(my_list).value_counts()


So, most of our research articles have either 1 or 2 tags.

Data cleaning and preprocessing for OneVsRest Classifier

Before proceeding with data cleaning and pre-processing, it’s a good idea to first print and observe some random samples from training data in order to get an overview. Based on my observation I built the below pipeline for cleaning and pre-processing the text data:

De-contraction → Removing special chars → Removing stopwords →Stemming

First, we define some helper functions needed for text processing.

De-contracting the English phrases —

def decontracted(phrase): #specific phrase = re.sub(r”won’t”, “will not”, phrase) phrase = re.sub(r”can’t”, “cannot”, phrase) # general phrase = re.sub(r”n’t”, “ not”, phrase) phrase = re.sub(r”’re”, “ are”, phrase) phrase = re.sub(r”’s”, “ is”, phrase) phrase = re.sub(r”’d”, “ would”, phrase) phrase = re.sub(r”’ll”, “ will”, phrase) phrase = re.sub(r”’t”, “ not”, phrase) phrase = re.sub(r”’ve”, “ have”, phrase) phrase = re.sub(r”’m”, “ am”, phrase) phrase = re.sub(r”’em”, “ them”, phrase) return phrase

(I prefer my own custom set of stopwords to the in-built ones. It helps to me readily modify the stopwords set depending on the problem)

stopwords = [‘i’, ‘me’, ‘my’, ‘myself’, ‘we’, ‘our’, ‘ours’, ‘ourselves’, ‘you’, “you’re”, “you’ve”, “you’ll”, “you’d”, ‘your’, ‘yours’, ‘yourself’, ‘yourselves’, ‘he’, ‘him’, ‘his’, ‘himself’, ‘she’, “she’s”, ‘her’, ‘hers’, ‘herself’, ‘it’, “it’s”, ‘its’, ‘itself’, ‘they’, ‘them’, ‘their’, ‘theirs’, ‘themselves’, ‘what’, ‘which’, ‘who’, ‘whom’, ‘this’, ‘that’, “that’ll”, ‘these’, ‘those’, ‘am’, ‘is’, ‘are’, ‘was’, ‘were’, ‘be’, ‘been’, ‘being’, ‘have’, ‘has’, ‘had’, ‘having’, ‘do’, ‘does’, ‘did’, ‘doing’, ‘a’, ‘an’, ‘the’, ‘and’, ‘but’, ‘if’, ‘or’, ‘because’, ‘as’, ‘until’, ‘while’, ‘of’, ‘at’, ‘by’, ‘for’, ‘with’, ‘about’, ‘against’, ‘between’, ‘into’, ‘through’, ‘during’, ‘before’, ‘after’, ‘above’, ‘below’, ‘to’, ‘from’, ‘up’, ‘down’, ‘in’, ‘out’, ‘on’, ‘off’, ‘over’, ‘under’, ‘again’, ‘further’, ‘then’, ‘once’, ‘here’, ‘there’, ‘when’, ‘where’, ‘why’, ‘how’, ‘all’, ‘any’, ‘both’, ‘each’, ‘few’, ‘more’, ‘most’, ‘other’, ‘some’, ‘such’, ‘only’, ‘own’, ‘same’, ‘so’, ‘than’, ‘too’, ‘very’, ‘s’, ‘t’, ‘can’, ‘will’, ‘just’, ‘don’, “don’t”, ‘should’, “should’ve”, ‘now’, ‘d’, ‘ll’, ‘m’, ‘o’, ‘re’, ‘ve’, ‘y’, ‘ain’, ‘aren’, “aren’t”, ‘couldn’, “couldn’t”, ‘didn’, “didn’t”, ‘doesn’, “doesn’t”, ‘hadn’, “hadn’t”, ‘hasn’, “hasn’t”, ‘haven’, “haven’t”, ‘isn’, “isn’t”, ‘ma’, ‘mightn’, “mightn’t”, ‘mustn’, “mustn’t”, ‘needn’, “needn’t”, ‘shan’, “shan’t”, ‘shouldn’, “shouldn’t”, ‘wasn’, “wasn’t”, ‘weren’, “weren’t”, ‘won’, “won’t”, ‘wouldn’, “wouldn’t”]

Alternatively, you can directly import stopwords from word cloud API —

from wordcloud import WordCloud, STOPWORDS stopwords = set(list(STOPWORDS))

Stemming using Porter stemmer —

def stemming(sentence): token_words = word_tokenize(sentence) stem_sentence = [] for word in token_words: stemmer = PorterStemmer() stem_sentence.append(stemmer.stem(word)) stem_sentence.append(“ “) return “”.join(stem_sentence)

Now that we’ve defined all the functions, let’s write a text pre-processing pipeline —

def text_preprocessing(text): preprocessed_abstract = [] for sentence in text: sent = decontracted(sentence) sent = re.sub(‘[^A-Za-z0–9]+’, ‘ ‘, sent) sent = ‘ ‘.join(e.lower() for e in sent.split() if e.lower() not in stopwords) sent = stemming(sent) preprocessed_abstract.append(sent.strip()) return preprocessed_abstract

Preprocessing the train data abstract text—

train_data[‘preprocessed_abstract’] = text_preprocessing(train_data[‘ABSTRACT’].values) train_data[[‘ABSTRACT’, ‘preprocessed_abstract’]].head()


Likewise, preprocessing the test dataset –

test_data[‘preprocessed_abstract’] = text_preprocessing(test_data[‘ABSTRACT’].values) test_data[[‘ABSTRACT’, ‘preprocessed_abstract’]].head()


Now we longer need the original ‘ABSTRACT’ column. You may drop this column from the datasets.

Text data encoding

Splitting train data into train and validation datasets —

X = train_data[[‘Computer Science’, ‘Mathematics’, ‘Physics’, ‘Statistics’, ‘preprocessed_abstract’]] y = train_data[target_cols] from sklearn.model_selection import train_test_split X_train, X_cv, y_train, y_cv = train_test_split(X, y, test_size = 0.25, random_state = 21) print(X_train.shape, y_train.shape) print(X_cv.shape, y_cv.shape)


As we can see, we have got ~ 10500 datapoints in our training set and ~3500 datapoints in the validation set.

TF-IDF vectorization of text data

Building vocabulary —

combined_vocab = list(train_data[‘preprocessed_abstract’]) + list(test_data[‘preprocessed_abstract’])

Yes, here I’ve knowingly committed a sin! I have used the complete train and test data for building vocabulary to train a model on it. Ideally, your model shouldn’t be seeing the test data.

vectorizer = TfidfVectorizer(min_df = 5, max_df = 0.5, sublinear_tf = True, ngram_range = (1, 1)) X_train_tfidf = vectorizer.transform(X_train[‘preprocessed_abstract’]) X_cv_tfidf = vectorizer.transform(X_cv[‘preprocessed_abstract’]) print(X_train_tfidf.shape, y_train.shape) print(X_cv_tfidf.shape, y_cv.shape)


After TF-IDF encoding we obtain 9136 features, each of them corresponding to a distinct word in the vocabulary.

Some important things you should know here —

I didn’t directly jump to a conclusion that I should go with TF-IDF vectorization. I tried different methods like BOW, W2V using a pre-trained GloVe model, etc. Among them, TF-IDF turned out to be the best performing so here I’m demonstrating only this.

It didn’t magically appear to me that I should be going with uni-grams. I tried bi-grams, tri-grams, and even four-grams; the model employing the unigrams gave the best performance among all.

Text data encoding is a tricky thing. Especially in competitions where even a difference of 0.001 in the performance metric can push you several places behind on the leaderboard. So, one should be open to trying different permutations & combinations at a rudimentary stage.

Before we proceed with modeling, we stack all the features(topic features + TF-IDF encoded text features) together for both train and test datasets respectively.

from scipy.sparse import hstack X_train_data_tfidf = hstack((X_train[topic_cols], X_train_tfidf)) X_cv_data_tfidf = hstack((X_cv[topic_cols], X_cv_tfidf)) Multi-label classification using OneVsRest Classifier

Until now we were only dealing with refining and vectorizing the feature variables. As we know, this is a multi-label classification problem and each document may have one or more predefined tags simultaneously. We already saw that several datapoints have 2 or 3 tags.

Most traditional machine learning algorithms are developed for single-label classification problems. Therefore a lot of approaches in the literature transform the multi-label problem into multiple single-label problems so that the existing single-label algorithms can be used.

(‘C’ denotes inverse of regularization strength. Smaller values specify stronger regularization).

from sklearn.multiclass import OneVsRestClassifier from sklearn.linear_model import LogisticRegression C_range = [0.01, 0.1, 1, 10, 100] for i in C_range: clf = OneVsRestClassifier(LogisticRegression(C = i, solver = ‘sag’)), y_train) y_pred_train = clf.predict(X_train_data_tfidf) y_pred_cv = clf.predict(X_cv_data_tfidf) f1_score_train = f1_score(y_train, y_pred_train, average = ‘micro’) f1_score_cv = f1_score(y_cv, y_pred_cv, average = ‘micro’) print(“C:”, i, “Train Score:”,f1_score_train, “CV Score:”, f1_score_cv) print(“- “*50)


We can see that the highest validation score is obtained at C = 10. But the training score here is also very high, which was kind of expected.

Let’s tune the hyper-parameter even further —

from sklearn.multiclass import OneVsRestClassifier from sklearn.linear_model import LogisticRegressionC_range = [10, 20, 40, 70, 100] for i in C_range: clf = OneVsRestClassifier(LogisticRegression(C = i, solver = ‘sag’)), y_train) y_pred_train = clf.predict(X_train_data_tfidf) y_pred_cv = clf.predict(X_cv_data_tfidf) f1_score_train = f1_score(y_train, y_pred_train, average = ‘micro’) f1_score_cv = f1_score(y_cv, y_pred_cv, average = ‘micro’) print(“C:”, i, “Train Score:”,f1_score_train, “CV Score:”, f1_score_cv) print(“- “*50)


The model with C = 20 gives the best score on the validation set. So, going further, we take C = 20.

If you notice, here we have used the default L2 penalty for regularization as the model with L2 gave me the best result among L1, L2, and elastic-net mixing.

Determining the right thresholds for OneVsRest Classifier

The default threshold in binary classification algorithms is 0.5. But this may not be the best threshold given the data and the performance metrics that we intend to maximize. As we know, the F1 score is given by —

A good threshold(for each distinct label) would be the one that maximizes the F1 score.

def get_best_thresholds(true, pred): thresholds = [i/100 for i in range(100)] best_thresholds = [] for idx in range(25): best_thresh = thresholds[np.argmax(f1_scores)] best_thresholds.append(best_thresh) return best_thresholds

In a nutshell, what the above function does is, for each of the 25 class labels, it computes the F1 scores corresponding to each of the hundred thresholds and then selects that threshold which returns the maximum F1 score for the given class label.

If the individual F1 score is high, the micro-average F1 will also be high. Let’s get the thresholds —

clf = OneVsRestClassifier(LogisticRegression(C = 20, solver = ‘sag’)), y_train) y_pred_train_proba = clf.predict_proba(X_train_data_tfidf) y_pred_cv_proba = clf.predict_proba(X_cv_data_tfidf) best_thresholds = get_best_thresholds(y_cv.values, y_pred_cv_proba) print(best_thresholds)


[0.45, 0.28, 0.19, 0.46, 0.24, 0.24, 0.24, 0.28, 0.22, 0.2, 0.22, 0.24, 0.24, 0.41, 0.32, 0.15, 0.21, 0.33, 0.33, 0.29, 0.16, 0.66, 0.33, 0.36, 0.4]

As you can see we have obtained a distinct threshold value for each class label. We’re going to use these same values in our final OneVsRest Classifier model. Making predictions using the above thresholds —

y_pred_cv = np.empty_like(y_pred_cv_proba)for i, thresh in enumerate(best_thresholds): print(f1_score(y_cv, y_pred_cv, average = ‘micro’))



Thus, we have managed to obtain a significantly better score using the variable thresholds.

So far we have performed hyper-parameter tuning on the validation set and managed to obtain the optimal hyperparameter (C = 20). Also, we tweaked the thresholds and obtained the right set of thresholds for which the F1 score is maximum.

Making a prediction on the test data using OneVsRest Classifier

Using the above parameters let’s move on to build train a full-fledged model on the entire training data and make a prediction on the test data.

# train and test data X_tr = train_data[[‘Computer Science’, ‘Mathematics’, ‘Physics’, ‘Statistics’, ‘preprocessed_abstract’]] y_tr = train_data[target_cols] X_te = test_data[[‘Computer Science’, ‘Mathematics’, ‘Physics’, ‘Statistics’, ‘preprocessed_abstract’]] # text data encoding X_tr_tfidf = vectorizer.transform(X_tr['preprocessed_abstract']) X_te_tfidf = vectorizer.transform(X_te['preprocessed_abstract']) # stacking X_tr_data_tfidf = hstack((X_tr[topic_cols], X_tr_tfidf)) X_te_data_tfidf = hstack((X_te[topic_cols], X_te_tfidf)) # modeling and making prediction with best thresholds clf = OneVsRestClassifier(LogisticRegression(C = 20)), y_tr) y_pred_tr_proba = clf.predict_proba(X_tr_data_tfidf) y_pred_te_proba = clf.predict_proba(X_te_data_tfidf) y_pred_te = np.empty_like(y_pred_te_proba) for i, thresh in enumerate(best_thresholds):

Once we obtain our test predictions, we attach them to the respective ids (as in the sample submission file) and make a submission in the designated format.

ss = pd.read_csv(‘SampleSubmission.csv’) ss[target_cols] = y_pred_te ss.to_csv(‘LR_tfidf10k_L2_C20.csv’, index = False)

The best thing about participating in the hackathons is that you get to experiment with different techniques, so when you encounter similar kind of problem in the future you have a fair understanding of what works and what doesn’t. And also you get to learn a lot from other participants by actively participating in the discussions.

You can find the complete code here on my GitHub profile.

About the Author


Psychological Research With Human Participants

Psychological science tries to figure out why humans and animals behave the way they do. Typically, psychologists describe behavior as visible actions such as eating, recalling tales, and so on. What about subliminal psychological processes like thinking and feeling?

Although thoughts and feelings are not physically visible, they impact behavior characteristics such as response time and blood pressure, which are frequently employed to quantify these intangible processes. The practical benefits of psychology research are many. They include discoveries such as improved techniques for treating psychologically ill persons, better vehicle designs to make them easier and safer to operate, and innovative means of improving worker performance and pleasure.

Psychological Research with Human Participants

Research using human beings has been extremely helpful in expanding our understanding of the biological, behavioral, and social sciences. Federal, state, and local authorities carefully restrict this type of study. Professional associations have also created discipline-specific norms, rules, and regulations to ensure study participants’ well-being and rights. The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research was established to investigate concerns regarding safeguarding people in research at the beginning of the 1970s in response to widely reported research abuse.

The Commission published the Ethical Principles and Guidelines for the Protection of Human Subjects of Research report, also known as the Belmont Report, in 1979 and established the ethical foundation for the present federal regulations that safeguard human respondents. The laws and rules that impact how research with human subjects is conducted are briefly listed below.


The Health Insurance Portability and Accountability Act (HIPAA) of 1996 is part of a new law that addresses privacy and security problems in particular domains such as electronic healthcare transactions. HIPAA applies to health plans, health care providers who send any health information in electronic form in conjunction with a HIPAA-covered transaction, and health care clearinghouses.

HIPAA’s scope, however, extends to outsourcers as well, as the Act requires covered entities to impose HIPAA obligations on entities that are business associates and deal with the covered entity and perform a function/service that involves the use of individually available health information of the covered entities and receive health information. HIPAA establishes two types of requirements: privacy standards and security standards. The privacy rule prohibits the disclosure or use of protected health information unless expressly approved by the individual or required by law. The security rule is a subset of the former and takes effect when protected health information is transferred or stored electronically.

Public Health Services Act

The Public Health Service Act is a federal law adopted in the United States in 1944. Title 42 of the United States Code (The Public Health and Welfare), Chapter 6A, contains the complete statute (Public Health Service). The legislation established the federal government’s quarantine jurisdiction for the first time. It charged the US Public Health Service with preventing the import, transfer, and spread of infectious illnesses from other nations into the US. The initial authorization for scientists and special consultants to be appointed “without regard to the civil-service regulations,” known as a Title 42 appointment, was provided by the Public Health Service Act.

National Institutes of Health (NIH)

The National Institutes of Health (NIH) is the principal federal organization in charge of conducting and financing biomedical and behavioral research. It has important functions in biomedical research training and health information dissemination. The National Institutes of Health aims to seek basic information about the nature and behavior of living systems and to use that knowledge to improve health, prolong life, and minimize sickness and disability.

The agency comprises the Office of the Director, in charge of general policy and program management, and 27 institutes and centers, each focusing on certain diseases or areas of human health research. A highly competitive system of peer-reviewed grants and contracts supports a diverse array of research.

Code of Conduct for Belmont Report

The Belmont Report highlighted three basic codes of conduct that ought to be undertaken by researchers as their foremost responsibilities, each of which has been elucidated below.

Respect for People

After receiving the above-described information about the study, survey respondents must opt to take part in it as a portion of the written consent procedure. As a result, participants may feel pressured to participate in the study, which presents a problem for them. To lessen any perceived coercion to engage, for instance, the rights of imprisoned individuals should be properly taken into account. This is also a problem if students are involved in a study when the instructor is also a researcher.

In this situation, the teacher must make it plain during the informed consent procedure that the participants’ ability to decline participation will not impact how well they perform in the course. A translation of the written consent documentation must be given if there are non-English speaking participants. If minors or people under legal guardianship are involved, the legal guardian’s informed consent and approval must be sought. The assent procedure must outline the requirements for participation and do it in a way that makes it clear to the participants what is expected of them.


Despite being a clear factor, physical danger is hardly mentioned in a few research. Threats to a participant’s psychological well-being, prestige, and social status are more frequent. Participants may experience stress or emotional distress as a result of certain studies. Participants in a study may suffer psychological injury if they are prompted to think about traumatic or unpleasant occurrences while in an interview or when responding to a questionnaire.

To assess mood states, depressive mood inducement may take place in some investigations. As a result, changing individuals’ moods could have a negative psychological impact. There may be a social risk that infants’ confidentiality is violated while distributing a research study. In that case, the researcher may protect the participants’ privacy at all times while the research is being conducted.

Therefore, investigations with a greater chance of discovering anything significant may also be more dangerous than those with a lower chance of success. However, the researcher is responsible for evaluating whether circumstances would be too detrimental for volunteers to participate in a study. Systematic abuse of study participants is never justifiable.


The justice principle includes fair participant screening. Both ensure that possibly hazardous circumstances are not only administered to a particular group and guarantee that everyone has an equal opportunity to receive potentially positive treatments in research (for example, treatments for particular mental illnesses or conditions). Special attention must be given to groups that might be simpler to control (e.g., people with illnesses and poor individuals).

Consider, for instance, that we are a researcher undertaking a study in a region with a high concentration of financially deprived people. We intend to pay the respondents $50 to participate in our in-depth study (we intend to conduct in-depth interviews with the participants and spend some time observing them).

It is reasonable to pay U.S. students $50 for this sort of participation, so we compensate poor attendees in our study with the same amount. $50, however, means something different to middle-class people than it does to low-income people. Even if the volunteers did not wish to take part in our research, they might feel pressured to do so to receive the  $50, which might be enough to cover their family’s expenses for a while.

Therefore, many view this kind of compensation as coercive toward low-income participants. Because of their more pressing need for remuneration than people with higher incomes, these people could feel they have fewer choices than those with higher incomes. To assure equitable participant selection, researchers need to consider these concerns. A scientific rationale must be provided if one participant group should be omitted from a research project.

Principles and their Application

These are −

Respect to Persons Beneficence Justice

Provides information about the study before it begins (nature of participation, purpose, risks, benefits) Reduces risk of harm to participants Selection of participants must be fair Obtains voluntary consent from participants after they are informed Potential benefits of the study must outweigh risks All participant groups must have opportunity to receive benefits of research

Gives participants opportunity to ask questions Inhumane treatment of participants is never justified No participant groups may be unfairly selected for harmful research

Informs participants of right to withdraw


Even though research on human participants provides a rich repertoire of information and knowledge that can facilitate a magnanimous sum of life, it is crucial to remember that morality and ethics come first. We must always be so blinded in trying to arrive at a conjecture that we remember the cardinal principle that we are using other human beings as a bridge. Therefore, various statutory bodies have established essential rules and regulations to keep all of this in check.

Update the detailed information about Issues Of Validity In Evolutionary Psychology Research on the website. We hope the article's content will meet your needs, and we will regularly update the information to provide you with the fastest and most accurate information. Have a great day!